Archive RiskBasedBook

Risk-Based Decision-Making for
Assessing Petroleum Impacts at

Exploration and Production Sites
Edited by:
Sara J. McMillen, Chevron Research and Technology Company
Renae I. Magaw, Chevron Research and Technology Company
Rebecca L. Carovillano, CH2M HILL
RISK-BASED DECISION-MAKING FOR ASSESSING

PETROLEUM IMPACTS AT EXPLORATION AND PRODUCTION SITES
EDITED BY: MCMILLEN, MAGAW, CAROVILLANO

Edited by:
THIS BOOK WAS PUBLISHED IN COOPERATION WITH

THE P ETROLEUM ENVIRONMENTAL R ESEARCH FORUM
& THE UNITED STATES DEPARTMENT OF ENERGY
ISBN 0-9717288-0-1
October 2001
Published by the Department of Energy and the Petroleum Environmental Research
Forum.
Printed by the Government Printing Office
Additional copies of this report are available from:
Nancy Comstock
U.S. Department of Energy
National Energy Technology Laboratory
National Petroleum Technology Office
One West Third Street
Tulsa, OK 74103-3519
All rights reserved. No part of this book may be reproduced, stored in a retrieval system,
or transcribed in any form or by any means, electronic or mechanical, including
photocopying and recording, without the prior written permission of the publisher.
Printed in the United States of America
Table of Contents

Table of Contents
Preface..........................................................................................................................1
Acknowledgments.......................................................................................................6
List of Abbreviations ..................................................................................................8
Executive Summary ..................................................................................................15
Chapter 1 - An Overview of Risk-Based Decision-Making
for Site Management.................................................................................................19
Chapter 2 - Technical Basis for Current Soil Management Levels of
Total Petroleum Hydrocarbons................................................................................36
Chapter 3 - Composition of Crude Oils and Gas Condensates .............................46
Chapter 4 - Application of the Total Petroleum Hydrocarbon
Criteria Working Group (TPHCWG) Methodology to Crude Oils and Gas
Condensates ..............................................................................................................58
Chapter 5 - Performance of TPHCWG Analytical Method for Crude
Oils in Soils................................................................................................................77
Chapter 6 - Predicting the Fate and Transport of Hydrocarbons in Soil
and Groundwater ......................................................................................................89
Chapter 7 - Development of Reference Doses for Heavy Total Petroleum
Hydrocarbon (TPH) Fractions................................................................................111
Chapter 8 - Developing Total Petroleum Hydrocarbon Risk-Based
Screening Levels for Sites Impacted by Crude Oils and Gas Condensates........121
Chapter 9 - Hydrocarbon Transport From Oil and Soil to Groundwater...........132
Chapter 10 - Risk-Based Soil Screening Levels for Crude Oil: The Role of
Polyaromatic Hydrocarbons...................................................................................142
Chapter 11 - An Evaluation of Benzene Risk.......................................................156
Chapter 12 - Risk Evaluation of Metals in Crude Oils ........................................174
Chapter 13 - Application of Risk-Based Decision-Making for
International Exploration and Production Site Management...............................187
Appendix..................................................................................................................198
Glossary of Terms .................................................................................................. 236
iii
Preface
Preface
Remarkable changes in the approach used for protection of human health and
the environment in the United States have occurred over the past 30 years.
During the period from the early 1970s through the early 1990s, setting
priorities on which environmental problems were important was not an issue.
The problems and apparent culprits were obvious. The public and regulatory
agencies agreed on the severity of environmental problems and what was needed
to correct them. As a result, legislation was passed, regulations were
promulgated, and efforts and resources were focused on obvious pollution
control problems. Meeting these regulatory needs became the primary goal of
municipalities and industries during this period.
Over time, the increasing number of environmental laws and regulations, as well
as the increasing complexity of environmental problems, began to strain the
resources of regulatory agencies, municipalities, and industry. Interest increased
in identifying and considering different approaches for environmental
protection. As early as the mid-1980s, it was recognized that approaches other
than the compliance, command, and control approach for environmental
protection were needed. While he was Administrator of the United States
Environmental Protection Agency (USEPA) in the mid-1980s, William
Ruckelshaus attempted to introduce the concepts of risk assessment and risk
management into the USEPA decision-making process. Attempts also were
made to distinguish between the two concepts.
In 1990, then USEPA Administrator William Reilly called for a national debate
on national directions and policies for environmental protection. What
stimulated that call was the 1990 USEPA Science Advisory Board Report,
Reducing Risk: Setting Priorities and Strategies for Environmental Protection.
Among the many recommendations in that report was one that recommended
targeting environmental protection efforts on the basis of environmental risk and
risk reduction opportunities. Also recommended was the use of risk-based
priorities in environmental planning and budgeting.
Winston Churchill, at a critical turning point in World War II, stated, This is
not the end. It is not even the beginning of the end. But it is, perhaps the end of
the beginning. Unrecognized at the time, but obvious in hindsight, is the fact
that, in the United States, about the year 1990 was the end of the beginning
period of environmental protection in which government laws and regulations
were considered as the best approach for such protection. From the early
1990s, environmental protection efforts increasingly have focused on risk
assessment and risk management approaches.
Preface
It is clear that the compliance, command, and control approach that dominated
regulatory enforcement policy since the 1970s is no longer the predominant
driver of the profession and the environmental industry. It also is clear that we
will see a continuing firm regulatory baseline, with strong enforcement to define
the floor for environmental progress and sustain the legal mechanism for
penalizing environmental violators. Thus, we will have both a strong regulatory
program and an increasing emphasis on risk assessment and risk management
approaches for environmental protection.
From the initial efforts in the 1980s, risk-based evaluations and decisions have
become an important component of efforts to protect human health and the
environment. The information in Chapter 1 of this document provides an
excellent summary of the development of the risk-based decision-making
(RBDM) process and the current state of its use, particularly for sites with
contaminated soils. Thus, Chapter 1 is an overview of the important concepts
and incremental growth of RBDM in this second and important phase of
environmental protection.
A value of the RBDM process is that each site is treated individually and that
the remedial measures result in cleanup levels that are environmentally
acceptable for the given site characteristics and anticipated land use. For
situations where there are groups of sites, the RBDM process allows resources
to be focused on sites or areas that have greater environmental concerns. The
RBDM process also allows regulations to incorporate reasonable but
conservative Tier 1 screening levels. This avoids the use of overly
conservative approaches such as achieving background concentrations or
utilizing technologies deemed best available by non-risk based criteria.
Utilization of the RBDM process:
Recognizes the differences of each site,

Encourages owner rather than regulatory -led activities,
Uses and integrates human and ecological risk-based knowledge, and
Provides a focus on achieving site-specific environmental protection.
For sites with contaminated soils, the RBDM process results in achieving an
environmentally protective endpoint, i.e., a concentration of a chemical in such
soils below which there is no expected adverse effect to human health and the
environment.
The increased emphasis on RBDM has required greater knowledge of factors
that affect chemical release, transport, exposure, and general availability in the
environment. In turn, this need triggered scientific, pilot scale, and field
evaluations that have provided such knowledge.
Preface
In terms of risk-based assessments, for sites with contaminated soils, some of

the important scientific and technical knowledge that has resulted in the past
10 years includes the following:
The fact that a chemical, such as a petroleum hydrocarbon, can be

measured in a soil says nothing about the actual risk that that chemical
may pose.
Not all chemicals in a soil are easily released and leachable.
It is possible to allow some chemicals to remain at a site and still be

protective of human health and the environment.
Reasonable information is available to evaluate site-specific risks of

chemicals in a soil.
It is not mass removal of chemicals at a site, but rather risk reduction

related to the expected use of the site that is important.
Some concentrations of anthropogenic chemicals in a soil may have

limited availability.
The site remediation question is not how much contamination can be

cleaned up, but rather how much contamination should be cleaned
up.
Information about the chemical release in a soil and chemical exposure

to a receptor is important in the RBDM process.
Natural attenuation of chemicals such as petroleum hydrocarbons does

occur in soil and, under certain conditions, can be an effective
management approach for residual hydrocarbons.
Aging and weathering of anthropogenic hydrocarbons in soil can result

in greater sequestering and less release and leachability of such
chemicals.
Detailed knowledge on many of these and other important points is incorporated

in the various chapters of this book. For the first time, relevant information
about the composition of crude oils and condensates, about the technical basis
for risk-based decisions for petroleum hydrocarbons, and about risk-based soil
screening levels for specific chemicals such as petroleum hydrocarbons,
polyaromatic hydrocarbons, volatile organic chemicals such as benzene, and
metals have been brought together in one document. Equally relevant is the
quality and integrated nature of the assembled information. Thus, this book
provides scientific and field knowledge that are important for risk-based
Preface
decisions for the management of hydrocarbon-impacted soil at oil and gas

industry exploration and production (E&P) sites.
The authors of the respective chapters are among the most knowledgeable in the
nation in the respective scientific, technical, and policy areas. The authors are to
be congratulated for having prepared a document that presents a large number of
complex concepts and evaluations in a readable and succinct manner. This is an
important and timely contribution that will have value to many parts of the
environmental profession.
As indicated in Chapter 1, the general framework for risk-based decisionmaking was developed largely in response to sites impacted by hazardous
materials. The framework has been refined over time and shown to be
applicable to many sites and situations. This book provides information that can
be used to apply the framework to oil and gas industry E&P sites. It should be
noted, however, that the information in the book has much broader application.
Specific information in the book has relevance, on a selected basis, to other sites
that have petroleum hydrocarbons of the nature discussed. Such sites include
Brownfield sites, some CERCLA and RCRA sites, and sites with contaminated
sediments. Thus, a reader should consider the information provided for use not
only at E&P sites, but also at other sites requiring possible remediation. It
would do a disservice to the material in the book and to the insights provided by
the authors if the information in the book were considered of relevance only to
E&P sites.
Also noted in Chapter 1 is the fact that many other industrialized nations, such
as Canada, New Zealand, and The Netherlands, use a risk-based approach for
developing regulations and assessing individual sites. Australia, Great Britain,
and Italy are among the other nations using or considering such an approach.
However, the concepts involved and the RDBM approach are equally applicable
to situations in developing countries. The knowledge that has been noted earlier
indicates that: a) it is possible to allow some petroleum hydrocarbons to remain
at a site and be protective of human health and the environment, and b) it is risk
reduction that is the most desirable environmental goal, not achieving generic
hydrocarbon concentration limits. The information in this book can be used by
developing countries as they formulate environmental regulatory programs that
are protective of human health and the environment. By doing so, developing
countries can benefit from the comprehensive knowledge generated in the last
decade. They thus will be at the leading edge of efforts to protect human health
and the environment while at the same time meeting other social and economic
needs.
As a final point, when considering the RBDM process, an important aspect is
that of risk communication. The potential impact of chemicals at a site,
knowledge of the uncertainties involved, the pathways of possible concern, and
Preface
analytical issues need to be discussed clearly and in a risk context. This book
provides such information in a logical and understandable manner. As such, it is
an important risk assessment and risk communication tool. It is my expectation
that future evaluations of E&P sites, and of other relevant sites, will be much
improved as a result of the information and logic presented in this book.
Raymond C. Loehr
H. M. Alharthy Centennial Chair
Environmental and Water
Resources Engineering Program
University of Texas at Austin
June 15, 2001
Acknowledgments
Acknowledgments
This book was prepared in conjunction with Petroleum Environmental Research
Forum (PERF) Project 97-08. The companies and trade associations that
supported this project include the American Petroleum Institute, Arthur D.
Little, the Canadian Association of Petroleum Producers, Chevron Research and
Technology Company, Equilon Enterprises LLC, ExxonMobil Corporation, Gas
Technology Institute, Texaco Incorporated, Unocal Corporation, and the United
States Department of Energy (DOE). Funding for this book was provided
through a grant from DOE/National Energy Technology Laboratory, under the
direction of Nancy Comstock of the National Petroleum Technology Office.
DOE provided research through its Lawrence Berkeley National Laboratory and
Pacific Northwest National Laboratory. Additional funding was provided to Dr.
William Rixey by the New Mexico Oil and Gas Association Total Petroleum
Hydrocarbon Working Group, chaired by Paul Peacock of Marathon Oil.
The authors and editors of this book include:

George Deeley, Equilon Enterprises LLC
Gregory Douglas, Arthur D. Little
Deborah A. Edwards, ExxonMobil Biomedical Sciences
Michele Emerson, Brown and Caldwell
William R. Gala, Chevron Research and Technology Company
Wayne A. Hamilton, Shell Exploration and Production Company
Jill M. Kerr, ExxonMobil Production Company
Raymond C. Loehr, University of Texas
H. Rodger Melton, ExxonMobil Upstream Research Company
David V. Nakles, ThermoRetec Consulting Corporation
George Naughton, Arthur D. Little
Kirk T. OReilly, Chevron Research and Technology Company
Ileana Rhodes, Equilon Enterprises LLC
William G. Rixey, University of Houston
H. James Sewell, Shell Exploration and Production Company
Lynn R. Spence, Spence Environmental Engineering
Robert E. Sweeney, Unocal Corporation
John H. Trefry, Florida Institute of Technology
Robert P. Trocine, Florida Institute of Technology
Ann Tveit, ExxonMobil Biomedical Sciences
Acknowledgments
In addition we would like to acknowledge the following individuals for their

contributions to PERF 97-08:
Rene F. Bernier, Texaco Incorporated

Veronica R. Blackwell, Chevron Research and Technology Company
Lin Callow, Gulf Canada
Karen DeLong, Chevron Business and Real Estate Services
George DeVaull, Equilon Enterprises LLC
Steven Geiger, ThermoRetec Consulting Corporation
Maged Hamed, Exxon Mobil Corporation
John Harju, Gas Technology Institute
Steven B. Hawthorne, University of North Dakota
Hoi-Ying Holman, Lawrence Berkeley National Laboratory
Harley Hopkins, American Petroleum Institute
Michael H. Huesemann, Pacific Northwest National Laboratory
Paul Lundegard, Unocal Corporation
Evan R. Sedlock, Chevron Research and Technology Company
Karen Synowiec, Chevron Research and Technology Company
List of Abbreviations
gw
Groundwater mixing zone thickness
Degradation constant in the saturated zone
unsat
Degradation constant in the unsaturated zone beneath the

contaminant source
Volatilization rate constant for the contaminant source zone
Soil bulk density
Soil particle density
Air filled porosity
as
Volumetric air content
Volumetric oil content in vadose zone soil
org
Porosity filled with residual phase
Water filled porosity
ws
Volumetric water content
A
ADEC
Cross-sectional area of the source

Alaska Department of Environmental Conservation
ADL
Average dose level for an exposure pathway
Ag
API
Silver
American Petroleum Institute
As
ASTM
Arsenic
American Society for Testing and Materials
AT
Averaging time
Atm
Ba
Atmosphere
Barium
BCE
Be
British Columbia Environment

Beryllium
BTEX
Benzene, toluene, ethylbenzene, xylenes
BW
C
Body weight
Chemical concentration
Cal-EPA
CCME
State of California Environmental Protection Agency

Canadian Council of Ministers of the Environment
Cd
Cadmium
CERCLA
CF
Comprehensive Environmental Response and Contingency

Liability Act
Calibration factor
Co
COC
Concentration in organic phase

Chemical of concern
Cof
Co max
Final concentration of chemical in the oil phase

Maximum concentration of chemical in the oil
Cppmv
Concentration in vapor reported as parts per million volume
Cr
CR
Chromium
Contact rate
Cs
CSF
Concentration of chemical sorbed to soil

Oral cancer slope factor
CSM
Conceptual site model
CT
CTPH
Total concentration of chemical in soil

Total concentration of TPH in soil
Cv
Cvf
Cvs
Concentration of chemical in vapor phase

Vapor concentration in soil outside of building structure or at
ground surface
Vapor concentration at the source of contamination
Cw
Concentration of chemical in water
Cw max
Cu
Maximum dissolved concentration

Copper
CVAAS
D
Cold-vapor atomic absorption spectrometry

Depth of source of contamination
DAFgw
Dilution attenuation factor in groundwater
DAFmix
DAFo
Dilution attenuation factor in the groundwater mixing zone

Overall dilution attenuation factor
DAFsat
Dilution attenuation factor in groundwater downstream of the

contaminant source
DAFunsat
Dilution attenuation factor in unsaturated zone
Dair
Deff
Molecular diffusion coefficient for chemical in air

Effective diffusion coefficient in soil
Dwater
Molecular diffusion coefficient for chemical in water
DwT,S
Total effective diffusion coefficient
DRO
E&P
Diesel range organics

Exploration and Production
EC
ED
Equivalent carbon
Exposure duration
EF
Exposure frequency
EPC
FAAS
Electronic pressure control

Flame atomic absorption spectrometry
FID
Foc
Flame-ionizing detector
Fraction of organic carbon in soil
Foil
FT
Fraction of oil in soil

Fate and transport
GC
Gas chromatograph or gas chromatography
GFAAS
GRO
Graphite furnace atomic absorption spectrometry

Gasoline range organics
GW
GW
GW Ing
Groundwater
Protective concentration limit for ingestion of groundwater
H
Hg
Henrys Law constant or distance from bottom of contaminant

source zone to the water table
Mercury
HI
HQ
Hazard Index
Hazard Quotient
HMPW
HSW
High melting point wax

High sulfur wax
Infiltration rate
ICP-MS
ID
Inductively coupled plasma-mass spectrometry

Inner diameter
IRsoil
IRw
Soil ingestion rate

Water ingestion rate
Partition coefficient
Kd
KD
Soil/water partition coefficient

Kuderna-Danish
KH
Henrys Law coefficient
10
Ko
Oil-water partition coefficient
Koc
Koil
Partition coefficient for soil organic carbon

Partition coefficient for oil
Kow
Ks
Octanol-water partition coefficient

Soil-water sorption coefficient
KT,S
Overall soil-water partition coefficient for the contaminant source

zone
KT, unsat
Overall soil-water partition coefficient for the unsaturated zone
Length of source area parallel to groundwater flow direction, or

volatilization distance
LA
LCS
Louisiana
Laboratory control samples
LDF
Lateral dilution factor
LF
LFSW
Leaching factor
Leaching factor from soil to water
LOAEL
M
Lowest-observed-adverse-effect level
Soil-to-skin adherence factor
MADEP
MCL
Massachusetts Department of Environmental Protection

Maximum contaminant level
MDL
Method detection limit
MF
mg/kg
Modifying factor
Milligrams per kilogram
mg/L
MGP
Milligrams per liter

Manufactured gas plants
Mo
Molybdenum
mol
MS
Mole
Mass spectrometry, or matrix spike
MSD
MSDS
Matrix spike duplicates

Material safety data sheet
M org
Mass of chemical in organic phase
Ms
MT
Mass of chemical sorbed to soil

Total mass of chemical in soil
Mv
Mw
Mass of chemical in vapor phase

Mass of chemical dissolved in groundwater
11
mV
Millivolts
MW
MW i
Molecular weight
Molecular weight of contaminant of interest
MW o
MW TPH
Average molecular weight of oil

Average molecular weight of TPH mixture
NA
Not available or not applicable
NAPL
ND
Non-aqueous phase liquid

Non-detect
Ni
NOAEL
Nickel
No-observed-adverse-effect level
NOW
O&G
Non-hazardous oilfield waste

Oil and grease
ORNL
Oak Ridge National Laboratory
OSHA
P
Occupational Safety and Health Administration

Pressure
PAH
Pb
Polyaromatic hydrocarbon
Lead
PCL
Protective concentration level
PERF
PF
Petroleum Environmental Research Forum

Partition factor
PHC
ppb
Petroleum hydrocarbon
Part per billion
ppm
Part per million
PQL
PRG
Practical quantitation limit

Preliminary Remediation Goal
Q
QC
Contaminant mass flux

Quality control
Correlation coefficient, universal gas constant, or retardation

factor
RAFd
Dermal relative absorption factor
RAFo
RBC
Oral relative absorption factor

Risk-based concentration
RBCA
Risk-based corrective action
12
RBDM
Risk-based decision making
RBSL
RBSLS-GW
Risk-based screening level

Risk-based screening level that is protective of groundwater
RCRA
RES
Resource Conservation and Recovery Act

Residual saturation
RfC
Reference concentration
RfD
RfDo
Reference dose
Oral chronic reference dose
RIVM
RPD
Dutch National Institute of Public Health and the Environment

Relative percent difference
RSD
RT
Relative standard deviation

Retention time
Solubility
SA
Sb
Skin surface area

Antimony
SCC
Se
Soil contamination concentration

Selenium
Seff
Effective solubility
SFo
Si
Oral cancer slope factor

Solubility of the compound of interest
SIM
SJV
Selected ion monitoring

San Joaquin Valley
Sn
Tin
SOC
SPE
Soil organic content

Society of Petroleum Engineers
SQG
SRM
Soil quality guidelines

Standard reference materials
SSL
SSTL
Soil screening level

Site-specific target level
Temperature
THQ
TNRCC
Target hazard quotient

Texas Natural Resource Conservation Commission
TOC
Total organic carbon
13
TPH
Total petroleum hydrocarbons
TPHCWG
TR
Total Petroleum Hydrocarbon Criteria Working Group

Target excess individual lifetime cancer risk
TRRP
TX
Texas Risk Reduction Program

Texas
Infiltration rate
UCM
UF
Unresolved complex mixture

Uncertainty factor
Ugw
USEPA
Groundwater Darcy velocity

United States Environmental Protection Agency
USGS
UST
United States Geological Survey

Underground storage tank
Vanadium
v
VF
Seepage velocity
Volatilization factor
VFp
VFs s
Ambient air partition factor for particulates

Ambient air partition factor for vapors
VP
Vapor pressure
W
X
Weight fraction
Mole fraction
Xmax
Maximum mole fraction
14
Executive Summary
Executive Summary
This book presents the scientific background material necessary to support a
risk-based decision-making (RBDM) approach for managing hydrocarbonimpacted soil at oil and natural gas industry exploration and production (E&P)
sites . Much of the information presented in the various chapters was generated
as part of a multi-year joint industry Petroleum Environmental Research Forum
(PERF) project that focused on improving the technical basis for environmental
management of E&P sites . The general RBDM approach is largely based on
guidance originally developed by the United States Environmental Protection
Agency (USEPA). That framework was later refined by the American Society
for Testing and Materials (ASTM) for evaluating individual chemicals and the
Total Petroleum Hydrocarbon Criteria Working Group (TPHCWG) and PERF
for complex petroleum hydrocarbon mixtures . The information and approach
described in this report have recently been used in the development of risk-based
regulatory progra ms in several oil and natural gas producing states in the United
States .
Detailed information about the composition of crude oils and gas condensates is
summarized and the total petroleum hydrocarbon (TPH) fractions and individual
chemicals of potential concern in these mixtures are identified. The fate,
transport, and toxicity information necessary for estimating potential risks to
human health is then described and used to generate risk-based screening levels
(RBSLs ) that can be used in a Tier 1, or screening level, risk evaluation. These
RBSLs represent soil concentrations that are protective of human health and are
calculated using exposure equations recognized by USEPA as providing
conservative estimates (i.e., lower than necessary for the protection of human
health) of acceptable concentrations in soil. The calculated Tier 1 RBSLs can
be compared to regulatory criteria to determine whether the criteria are
protective of human health, or in instances where there are no regulatory criteria,
the RBSLs can be used as limiting concentrations for TPH or specific chemicals
of concern in soil.
TPH RBSLs for Complex

Mixtures of Petroleum Hydrocarbons
Tier 1 TPH RBSLs are determined for seventy crude oils that are representative
of the wide range of crude oil types produced around the world. TPH RBSLs
are also derived for fourteen gas condensates. These RBSLs are based on the
potential non-cancer health risks that these mixtures might pose and typical
exposure pathways that exist at E&P sites . The potential cancer risks are
addressed in a separate chemical-specific analysis. The primary focus is on
commercial and non-residential uses of the sites , which best represent
reasonably expected land uses for E&P sites . Based on these land uses, the
15
Executive Summary
exposure pathways of greatest concern are associated with direct human contact
with hydro carbon-impacted soil (i.e., soil ingestion, inhalation of soil particles,
and dermal contact). Leaching to groundwater and volatilization to outdoor air
are of lower concern for the complex mixtures as a whole.
The TPH RBSLs calculated for direct contact with soil impacted by this wide
variety of crude oils range from 35,000 to 67,000 mg/kg (3.5-6.7% by weight)
TPH for non-residential sites . TPH RBSLs for condensates range from 39,000
to 116,000 mg/kg (3.9-11.6% by weight) TPH. These values are significantly
greater than the TPH concentration of 10,000 mg/kg that has often been used as
the regulatory criterion for E&P sites . The TPH RBSLs for selected E&P
wastes are also determined for the same exposure scenario. These values were
very similar to those for the crude oils, ranging from 50,000 to 89,000 mg/kg
(5.0-8.9% by weight). Since these results represent RBSLs for all of the types
of oils produced around the world and because they have been shown to be
conservative for associated wastes and soil, they can be used to set TPH RBSLs
for crude oil and condensates present in soils at all E&P sites .
It is important to remember that these RBSLs apply to non-residential land use
conditions. If a particular site is expected to be used for residential purposes,
TPH RBSLs appropriate for that land use should be developed. Residential
RBSLs may be an order of magnitude or more lower than those appropriate for
non-residential land use.
Potential Health Risks of Metals,

Polyaromatic Hydrocarbons, and Benzene
In addition to the potential risks posed by the petroleum mixture as a whole,
which are addressed in terms of TPH, the potential risks posed by typical
indicator chemicals are also considered. The concentrations of metals in crude
oils are not sufficiently high to pose a significant health risk at residential sites
even at an overall oil concentration in excess of 650,000 mg/kg (65 wt %).
Therefore, metals are unlikely to be a major risk management consideration at
crude oil spill sites, and routine analyses for metals in soils at crude oil spill sites
is not recommended. However, it may be necessary to evaluate metals at those
sites where multiple spills may have occurred or at landfarms which have
received multiple applications of oily wastes .
The concentrations of polyaromatic hydrocarbons (PAHs) in crude oils and
condensates are not sufficiently high to pose a significant health risk for
residential land use, even at an overall oil concentration in excess of 14,000
mg/kg or (1.4 wt %). The only exception is naphthalene, a non-carcinogenic
PAH, which may pose a risk at lower concentrations. The concern relates to
naphthalenes potential for leaching to groundwater. Overall, these results
suggest that carcinogenic PAHs are unlikely to be a major risk management
16
Executive Summary
consideration at crude oil or condensate spill sites. Routine analysis for these
chemicals in soil at E&P sites is not necessary to ensure protection of human
health.
The understanding of the impact of benzene on the management of E&P sites is
continuing to evolve. Benzene is typically considered to be a chemical of
potential concern at gasoline release sites ; however, most upstream regulatory
programs do not routinely require benzene analysis of E&P site soils and do not
routinely set regulatory limits for benzene in soil. In this study, benzene was
detected in crude oils at a mean concentration of 1,340 mg/kg oil and in
condensates at a mean concentration of 10,300 mg/kg. However, some crude
oils contain little or no benzene. The potential for benzene to leach to
groundwater should be carefully evaluated at E&P sites where groundwater
protection is an important consideration and where the oil or condensate is
known to contain significant amounts of benzene. It may be appropriate to
directly analyze for benzene at the site (using USEPA 8260) and to derive
chemical-specific benzene RBSLs for groundwater protection. In deriving
appropriate benzene RBSLs, careful consideration should be given to relevant
environmental fate and transport processes, including volatilization and
biodegradation. It is now known that assumptions made in previous screening
level evaluations are overly conservative for benzene, because of its specific fate
and transport properties. Previously published benzene RBSLs using these
overly conservative assumptions may be 10 to 1000 times lower than necessary
to protect human health.
Chemicals of Concern for E&P Site Management

TPH has historically been the primary criterion for environmental management
at E&P sites. Although typical TPH regulatory limits used in the past have not
been health risk-based criteria, the work conducted to date demonstrates that
acceptable health risk-based TPH levels can be developed and that acceptable
criteria for non-residential sites are well above those historically used. This
indicates that TPH can continue to be an important criterion for health riskbased E&P site management programs. Measurements of bulk TPH using
conventional analytical methods can continue to be used to assess compliance at
most, if not all, E&P sites.
For crude oil and condensate spill sites, metals and PAHs are unlikely to exceed
health risk-based levels. Volatile aromatics such as benzene, toluene,
ethylbenzene, xylenes, and naphthalene may need to be evaluated on a site by
site basis .
17
Executive Summary
RBSLs for International Applications

The RBDM process is a scientifically defensible, flexible, and yet standard
process that can be used to develop RBSLs. The health risk-based concepts
have been developed ni North America and Europe, but they are equally
applicable to sites anywhere in the world. The RBSLs derived in this book are
generally based on exposure pathways and assumptions commonly applied in
the United States. Meaningful RBSLs for international locations can be
developed if adequate consideration is given to relevant exposure factors for the
new locations such as differences in lifestyle, climate, and local geology.
18
Chapter 1
An Overview of Risk-Based Decision-Making

for Site Management
Renae I. Magaw1 and David V. Nakles 2
1
2
Chevron Research and Technology Company

ThermoRetec Consulting Corporation
Introduction
Risk-based decision-making is the process of making environmental
management decisions based upon an assessment of the potential risks that
chemicals at a site may pose to human health and the environment. The United
States Environmental Protection Agency (USEPA) has developed a general
framework for risk-based decision making based upon human health
considerations, and has established general guidelines for determining what
constitutes acceptable risk to human health [1,2]. These guidelines can be used
to determine when some type of risk management action is required at a site.
Although the overall framework for risk-based decision-making was originally
developed for use at sites impacted by hazardous materials, in reality it is
equally applicable to all types of sites, including oil and gas industry exploration
and production (E&P) sites.
The general framework for risk-based decision-making was originally developed
by the USEPA largely in response to the requirements of the Comprehensive
Environmental Response and Contingency Liability Act of 1980 (CERCLA).
This framework has been refined over time and several tiered approaches to
risk-based decision-making have subsequently been developed. A major goal of
the framework is to make certain that management decisions for
environmentally impacted sites provide an adequate level of protection for
human health and the environment. Therefore, a health risk evaluation process
was developed and the overall risk characterization is used to guide site
management decisions.
Risk-Based Decision-Making at
Environmentally-Impacted Sites
Historically, regulatory programs in the United States have established
environmental management goals (i.e., cleanup levels) for chemicals of potential
concern at specific sites based on:
Background (or naturally occurring) chemical concentrations (for

example, those typically found in unaffected areas).
19
Chapter 1
Analytical detection limits.
Concentrations that may be attainable if the most aggressive

technologies were used for site remediation.
However, since none of these goals is directly tied to the actual risks posed by
the chemicals of concern, there is no way to determine whether or not these
goals actually protect human health and the environment. In addition, there is
no way to determine the cost/benefit associated with achieving the management
goals listed above, since the benefit of the action cannot be determined. Without
any knowledge of the benefit resulting from a given action, there is no way to
prioritize actions to focus them on those problems where the greatest potential
for risk reduction exists. This could conceivably result in a portion of the public
being left at risk, and in the misallocation of both technical and financial
resources.
In contrast, risk-based approaches to site management clearly describe the
potential health benefits that might result from a particular environmental
management decision. Consequently, the actions that are taken at a site can be
evaluated and prioritized based on the actual reduction in risk that would be
achieved, and technical and financial resources can be allocated appropriately.
Like all technical methodologies and protocols, risk-based decision-making is
not necessarily applicable to every situation at every site. For example, there
may be instances where a risk-based assessment concludes that total petroleum
hydrocarbon (TPH) concentrations at a specific site do not pose a health risk.
However, these same concentrations may produce unsightly conditions that are
not aesthetically acceptable for the current and/or future land use. In these
instances, common sense should be used to guide site management decisions.
It is also important to think carefully about the assumptions that are made when
using risk-based decision-making for site management. Since there tends to be
limited data available to conduct a risk-based evaluation of a site, there is
generally a need to make some basic assumptions during the risk evaluation.
Examples of assumptions that need to be made may include the toxicity of the
materials in question or the duration and extent of potential exposures. In every
risk evaluation, it is important to understand the sensitivity of the risk-based
decisions to the assumptions made in order to determine how robust the
evaluation is and the circumstances that might justify the use of different
assumptions. The greatest criticism of risk-based decision-making in site
management is that it can be manipulated to produce any result that is desired by
the user. The primary defense to this criticism is to make certain that all
assumptions are technically justifiable, and to examine the sensitivity of the
outcome to the more critical of these assumptions.
20
Chapter 1
The Risk Evaluation Process

The risk evaluation process, as developed by the USEPA [2], involves four
elements:
Hazard Identification
Exposure Assessment
Toxicity (or Dose-Response) Assessment
Risk Characterization
Simply stated, a quantitative risk evaluation involves identifying the chemicals

of potential concern at a site, simulating their release and movement in the
environment, estimating their uptake by both human and environmental
receptors (a receptor is an exposed person, animal, or plant), and predicting the
potential health effects of the exposure. Each of the technical elements of the
risk evaluation process is described below.
Hazard identification is accomplished by collecting and reviewing site
assessment data and identifying the chemicals of potential concern and the
environmental media (e.g., soil, groundwater, and air) in which they can be
found. It answers the question What are the potential hazards at the site?
Exposure Assessment
The exposure assessment answers the question To how much of the chemicals
of potential concern are receptors exposed? The exposure assessment is an
extremely important part of the risk evaluation process because it introduces
site-specific factors into the characterization of the site risk. The exposure
assessment can be thought of as a three-step process in which: (1) the site
setting, which depicts the relative locations of the hazards and potential
receptors, is characterized, (2) complete exposure pathways are identified, and
(3) the magnitude of the potential exposure is estimated.
Characterizing the site setting identifies the receptors that might be exposed to
the chemicals of potential concern. A key issue in identifying these receptors is
the current and reasonably expected future land use for the site. Historically,
regulatory agencies have required site managers to consider all potential future
land uses, including residential use, in all risk analyses. This is not a reasonable
assumption for most E&P sites where more realistic future land uses include
ranch land, agricultural land, or park land. More recently, regulatory agencies in
the United States have focused on protecting current land uses and have allowed
more flexibility in the selection of appropriate future land use scenarios. This
has resulted in more flexibility in developing regulatory criteria for site cleanup.
21
Chapter 1
Once the receptors have been identified, the next step is to determine how they
might be exposed. This is a process in which potentially complete exposure
pathways are identified. In identifying these complete exposure pathways, the
sources of the chemicals at the site are determined and the ways in which they
may move around in the environment and be transported to places at which
receptors might be exposed are considered. For example, if a crude oil is spilled
on soil at a site, a worker in the area may be exposed by direct skin contact with
the impacted soil. In addition, some of the components of the crude oil may
volatilize into air and be inhaled by the worker, or some components may
migrate through the soil into the groundwater, be transported to a drinking water
well at some distance from the site, and subsequently be ingested. The final step
of the exposure assessment is to quantify the potential exposure to receptors
using standardized intake equations.
Toxicity (Dose-Response) Assessment
The toxicity assessment answers the question What dose levels of the
chemicals of potential concern may produce adverse health effects in people or
other receptors? In the toxicity assessment, chemicals are usually evaluated
separately for their abilities to cause cancer and other adverse health effects. All
chemicals have the inherent ability to cause adverse health effects at some dose
level, but only certain chemicals have the ability to cause cancer. Most of the
available toxicological data for both carcinogenic and non-carcinogenic
chemicals have been generated in the laboratory using pure chemicals that have
been added to the food or water of rats or mice. One of the major challenges
associated with the use of these data is in extrapolating the results for individual
chemicals to situations in which mixtures of chemicals, such as crude oil, may
be of concern. A second challenge is in extrapolating the laboratory results
obtained in rodents treated with chemicals to situations in which people are
exposed to chemicals. In both cases, the USEPA includes uncertainty factors
into the analysis to make certain that the toxicity of the chemical or mixture of
chemicals is not underestimated.
The final step of the risk evaluation for a site is one in which the results of the
exposure assessment are combined with the toxicity assessment to quantify the
potential risks to human health and the environment. The result is a
conservative risk estimate that is likely to overestimate the true risks posed by
the site. In reality, the true risk will most likely be much lower than the
estimated risk.
22
Chapter 1
Quantitative Human Health Risk Evaluation

As described above, the risk evaluation process is generally composed of
four elements:
Exposure Assessment
Toxicity (or Dose-Response) Assessment
Most often, this process is quantitative in nature, the result of which is a

numerical estimate of risk. The equations used to estimate risk are based on
those originally developed by USEPA [2]. The calculations and the default
assumptions that are commonly used in the equations are specifically
designed/selected to provide a result that is protective of human health.
Exposure Assessment: Calculation of Contaminant Intake
The quantitative exposure estimate determines the amount of chemical that is
taken in by a receptor for a given exposure route. The primary exposure routes
for humans are ingestion of groundwater, inhalation of air, dermal contact with
soil, and ingestion of soil. In all cases, the calculation of chemical intake
requires knowledge of:
The concentration of the chemical in the impacted medium, i.e.,

soil (mg/kg), air (g/m3), or water (mg/L).
The amount of the impacted medium that is taken in by the

receptor (i.e., liters of air or water or kilograms of soil).
The amount of the impacted medium that is taken in is determined by

identifying an exposure event, specifying the quantity of the medium that is
taken in per event, and specifying the frequency and duration of the event. The
intake is then converted to a dose level by dividing it by the body weight of the
receptor and averaging over an appropriate time period. This yields an average
daily dose or average lifetime daily dose expressed in mg/kg per day. The
averaging time period depends upon the health effect that is being addressed.
For example, the averaging time for carcinogenic effects is a lifetime of 70
years. On the other hand, for non-carcinogenic effects, the averaging time is
equal to the duration of the exposure (e.g., 25 years for an adult worker).
Chemical intake for an exposure pathway is determined using the following
equation:
23
Chapter 1
ADL =
C x CR EF x ED
x
BW
AT
where:
ADL = Average dose level (mg/kg BW-day)
C = Chemical concentration (e.g., mg/kg-soil or mg/L-water)
CR = Contact rate or the amount of impacted medium contacted per event
(e.g., liters/day, kg/day)
EF = Exposure frequency (days/year)
ED = Exposure duration (years)
BW = Body weight of the receptor (kg)
AT = Averaging time of the exposure (days)
Derivation of Toxicological Dose-Response Factors
In estimating risk, the exposure estimate is combined with a toxicological doseresponse factor. The dose-response factor depends upon the chemical, the route
of exposure, and the health effect that is of concern (i.e., carcinogenic or noncarcinogenic). It is generally derived by USEPA [13], or other regulatory
agencies, and is made available to the public for use by risk assessors. The data
on which these factors are based are usually generated in laboratory studies
using animals. The dose-response factors derived from these data include
reference doses (RfDs) or inhalation reference concentrations (RfCs) for
evaluating non-carcinogenic effects and cancer slope factors for evaluating
carcinogenic effects as described below:
Reference Doses (RfDs mg/kg-day): Estimate of daily

exposure that is likely to be without an appreciable risk of adverse
(non-carcinogenic) health effects during a lifetime of exposure.
Reference Concentrations (RfCs mg/m3): Estimate of a

continuous inhalation exposure to the human population that is
likely to be without an appreciable risk of adverse (noncarcinogenic) health effects during a lifetime.
Oral Cancer Slope Factor [CSF (mg/kg/day)-1]: Slope of the

relationship between the oral dose received by a receptor and the
carcinogenic response.
Calculation of Risk
The risk calculations for non-carcinogenic health effects are expressed in terms
of a unitless hazard quotient that is calculated using the following equation:
24
Chapter 1
Hazard Quotient=
AverageDaily Dose(mg/kg - day)

ReferenceDose(mg/kg - day)
The threshold level of acceptability for the Hazard Quotient that has been
established by the USEPA [1] is the value 1.0, although some state regulatory
agencies within the United States have established different levels of
acceptability. Hazard quotients greater than 1.0 typically require further site
analysis or some sort of site action.
The risk calculation for carcinogenic health effects is based on a somewhat
similar equation:
Risk = Average Lifetime Daily Dose (mg/kg-day) x Slope Factor (mg/kg-day)-1
This risk calculation also yields a unitless value. The acceptable individual
excess lifetime cancer risk range established by the USEPA [1] is 10-4 to 10-6
(one-in-ten thousand to one-in-one million excess cancer risk). Many state
regulatory agencies within the United States have established acceptable risk
target levels within this range.
Tiered Risk-Based Decision-Making Frameworks

One drawback of the risk-based decision-making process, as originally
developed by the USEPA, is that it can require a substantial investment of
technical and financial resources, as well as time. Also, the data required to
complete the risk evaluation are often not readily available. For these reasons,
tiered strategies tailored for specific types of sites have recently been developed
by United States regulatory agencies and by independent organizations to permit
its cost-effective use. One example of a tiered risk-based decision-making
framework is that developed by the American Society for Testing and Materials
(ASTM).
The first significant development by ASTM was the Standard Guide for RiskBased Corrective Action Applied at Petroleum Release Sites [3]. The
development of this guide was driven by the need to cost-effectively and
expeditiously manage underground storage tank (UST) sites. The guide was
finalized in 1995 and it has since been recognized by the USEPA and used by
many regulatory agencies in the United States to revise UST regulations.
ASTM completed a second guide in April 2000 with the development of the
Standard Guide for Risk -Based Corrective Action [4]. This effort expanded the
previous guide by facilitating the use of risk-based corrective action in United
States regulatory programs including voluntary cleanup programs, Brownfields
redevelopment, CERCLA, and Resource Conservation and Recovery Act
(RCRA) corrective action.
25
Chapter 1
In addition to these national efforts by ASTM, several state environmental

regulatory agencies within the United States have also initiated unified riskbased corrective action programs that include voluntary, CERCLA, and RCRA
corrective action programs. Examples of these programs are the Massachusetts
Contingency Plan [5], the Illinois Environmental Protection Agency Tiered
Approach to Corrective Action Objectives [6], Louisiana Department of
Environmental Quality Risk Evaluation/Corrective Action Program [7], and the
Texas Natural Resource Conservation Commission (TNRCC) Risk Reduction
Program [8].
Tier 1
Figure 1 illustrates the tiered risk-based decision-making approach described by
ASTM, which is commonly referred to as the Risk-Based Corrective Action
(RBCA) process [3]. In the first tier of the RBCA process, chemical
concentrations in soil may be compared to generic Tier 1 risk-based screening
levels (RBSLs). RBSLs are chemical-specific concentrations in environmental
media that are considered protective of human health. These screening levels
are often derived by state or federal regulatory agencies using very conservative
exposure assumptions. The USEPA has developed Tier 1 Soil Screening Levels
(SSLs) [9], Preliminary Remediation Goals (PRGs ) [10], Risk-Based
Concentrations (RBCs) [11], and Human Health Medium-Specific Screening
Levels [12] for evaluating potential human health effects at CERCLA sites.
These screening levels are similar to RBSLs in that they are risk-based
concentrations derived from standardized equations combining exposure
information assumptions with USEPA toxicity data.
26
Chapter 1
RISK-BASED CORRECTIVE ACTION PROCESS
INITIAL SITE ASSESSMENT

Collect general data on site
Emergency Response
Emergency
Response
Needed?
Yes
No
Exceed
Satisfy
TIER 1 ASSESSMENT
Compare contamination to generic RBSLs
Yes
No
Corrective
action
feasible?
TIER 2 ASSESSMENT
Refine study, generate site-specific SSTLs
Compare contamination to SSTLs
Exceed
Yes
Satisfy
No
Corrective
action
feasible?
Exceed
Final Corrective
Action
TIER 3 ASSESSMENT
Further refine study and SSTLs
Compare contamination to SSTLs
Satisfy
Assess need for ongoing

monitoring program
NO FURTHER ACTION
Figure 1. A RBCA flowchart illustrating tiers and decision points.

In addition to the screening levels developed by USEPA, some state programs
have developed Tier 1 RBSLs. For example, the TNRCC has developed
Protective Concentration Levels (PCLs) as Tier 1 RBSLs [8]. A PCL is defined
as the concentration of a chemical of concern which can remain within the
source medium and not result in levels which exceed the applicable human
health risk-based exposure limit or ecological protective concentration level at
the point of exposure for that exposure pathway [8].
As shown in Figure 1, tiered approaches generally start with an initial screening
stage, Tier 1, that uses a basic set of site assessment data and involves a
comparison of the concentrations of chemicals in the different environmental
media to predetermined Tier 1 RBSLs. If site concentrations are below the
Tier 1 RBSL, the conclusion is drawn that chemicals of potential concern do not
27
Chapter 1
pose a significant risk to human health or the environment and that no remedial
action is necessary. If site concentrations exceed Tier 1 levels, the site manager
generally has the option of remediating the site to Tier 1 levels or, alternatively,
progressing to a more data and labor intensive Tier 2 or even Tier 3 analysis.
Tiers 2 and 3
Tier 2 and Tier 3 analyses generally require increasingly sophisticated levels of
data collection and analysis, which in turn result in increased costs. The tradeoff for these increased costs will generally lie in lower remediation and overall
project costs, because the cleanup goals defined by a Tier 2 or 3 analysis are
likely to be higher than Tier 1 levels, and thus less costly to achieve. The
cleanup goals of the Tier 2 and 3 analyses are generally higher than the Tier 1
analysis because the generic assumptions used in the Tier 1 levels are replaced
with more relevant site-specific assumptions and data. They are not higher
because they are less protective of human health or the environment. In fact, all
three tiers of risk analysis provide an equal level of health protection. Tier 2 and
3 risk-based concentrations are often referred to as Site-Specific Target Levels
(SSTLs). ASTM [3,4] defines an SSTL as a risk-based remedial action target
level for chemical(s) of concern developed for a particular site under the Tier 2
and Tier 3 evaluations.
Upon completion of each tier, the site manager reviews the results and
recommendations, and decides if the cost of conducting the additional sitespecific analyses is warranted. Using the tiered approach, an E&P site manager
has the flexibility to forego the detailed risk characterization effort of a sitespecific Tier 2 or 3 analysis and proceed directly to site actions that generally
involve meeting conservatively low, generic site cleanup goals. In some cases,
this approach may be the more cost-effective and more prudent site management
decision.
Implementation of a Tiered Risk Approach
The development of tiered approaches for the risk-based analysis of sites was
based on the premise that there are situations where conducting a detailed risk
analysis may require more effort and time than immediate implementation of
site remedial actions. For this reason, after every tier of risk analysis, the site
manager must perform a cost/benefit evaluation to determine if it makes sense to
proceed to the next level of risk analysis. Only if a clear benefit exists would
the decision to move forward be made. For example, because the Tier 1
assessment is often based upon conservatively low, generic site cleanup goals,
the extent of a site remedial action may be larger (and more expensive) than
might be required if a more detailed site-specific Tier 2 analysis were
conducted. However, additional time and expense will be incurred to complete
the Tier 2 analysis. At this point the site manager must evaluate the potential
reduction in site remedial costs that may be realized by conducting the Tier 2
28
Chapter 1
analysis and compare that reduction to the additional cost of conducting the risk
analysis. If the potential savings outweigh the potential cost, it would be in the
managers best interest to move forward with the Tier 2 analysis. In some cases,
it is not the cost that drives the decision, but the schedule. If the time required to
conduct the next tier of risk analysis is not acceptable to regulatory agencies or
the public, then the decision to proceed with site remediation is essentially
made.
The decision to use the tiered risk-based strategies for site management is
usually dictated by the nature of the site contamination and the complexity of
the site conditions. However, it may also be dictated by the governing
regulatory agency, which may or may not accept the use of a tiered approach. At
most E&P sites, it is likely that a tiered risk-based decision-making strategy will
be the approach of choice. This is because E&P sites generally involve a known
and very limited number of chemicals of potential concern (e.g., crude oil, gas
condensates, and selected additives), and they have relatively small and simple
operational footprints. Consequently, the lower tiers of risk analysis will often
provide the most cost-effective site management approach.
Role of Generic Site Cleanup Criteria
in the Risk-Based Decision-Making Process
As described in Chapter 2, most regulatory pro grams, including those with
jurisdiction over E&P sites, have historically incorporated cleanup criteria that
are not explicitly health risk-based levels. Instead, generic criteria have often
been used, such as the commonly applied 1% TPH management level. For
practical purposes, these generic criteria may be used as Tier 1 screening level
criteria in a pseudo risk-based decision-making process. However, it should be
recognized that those generic criteria that are not risk-based may or may not be
protective of human health and the environment. One of the goals of a recent
joint-industry research project sponsored by the Petroleum Environmental
Research Forum was to derive generic risk-based screening criteria for E&P
sites that could be compared to existing, non risk-based criteria currently used
for E&P site management. See Chapter 8 for more information on RBSLs for
TPH.
The Derivation and Use of Risk-Based Screening Levels

RBSLs (chemical-specific concentrations in environmental media that are
considered protective of human health) can be derived from the risk equations
by specifying an acceptable target risk level. The equations are then rearranged
to determine the chemical concentration in the environmental medium of
concern that represents this risk level.
The tiered risk-based decision-making approach developed by ASTM relies
more on the use of RBSLs for site management decision-making, rather than on
29
Chapter 1
an explicit calculation of site risk, as generally used in the classical approach by

USEPA. The equation below is based on information provided by ASTM [3],
and is used to calculate RBSLs for non-carcinogenic health effects for the
exposure pathways of soil ingestion and dermal contact with soil:
mg
RBSL ss
=
kg/Soil
THQ x BW x ATn x 365
days
year
- 6 kg
10 mg x (IR soil x RAFo + SA x M x RAFd )
EF x ED
RfD o
where:
THQ
BW
ATn
EF
ED
IRsoil
RfDo
RAFd
RAFo
SA
M
=
=
=
=
=
=
=
=
=
=
=
Target hazard quotient for individual constituents (unitless)

Body weight (kg)
Averaging time for non-carcinogens (years)
Exposure frequency (days/year)
Exposure duration (years)
Soil ingestion rate (mg/day)
Oral chronic reference dose (mg/kg-day)
Dermal relative absorption factor (unitless)
Oral relative absorption factor (unitless)
Skin surface area (cm2/day)
Soil-to-skin adherence factor (mg/cm2)
Similar RBSL equations for other exposure pathways (i.e., inhalation of

volatiles and particulates, ingestion of groundwater, and inhalation of vapor) are
provided in the appendices of the ASTM guides [3,4]. It should be noted that
the derivation of RBSL equations that are appropriate for use with complex
mixtures, such as crude oil, requires additional manipulations of the equations
used above.
There are several factors in the risk equations that address the availability of
soil-bound contaminants to the human receptor. These are the dermal relative
absorption factor, or RAFd, the oral relative absorption factor, RAFo, and the
ambient air partition factors for both particulates (VFp) and vapors (VF ss).
These factors are included because the soil tends to bind many of the
contaminants and prevent them from coming into contact with the receptor and
causing an impact. This matrix effect is caused by the soil, is recognized by the
USEPA, and is currently the subject of a great deal of research by universities,
industry consortia, and the USEPA.
For the vapor and groundwater exposure pathways, the acceptable contaminant
concentrations in the air and groundwater can be used to back-calculate a
30
Chapter 1
contaminant concentration in soil that will be protective of these other media. In

other words, a contaminant concentration in soil can be determined that will not
result in an exceedance of the acceptable contaminant concentrations in air or
groundwater. To complete these back-calculations, a volatilization factor (VF,
[mg/m3]/ [mg/kg]) and leaching factor (LF, [mg/L]/[mg/kg]) are required. The
former predicts the amount of contaminant that will partition between the soil
and the vapor, while the latter predicts the partitioning from the soil to the
aqueous phase [3].
Other parameters required for the calculation of RBSLs include body weight,
exposure frequency, exposure duration, soil and water ingestion rates, air
inhalation rates, averaging times for carcinogens and non-carcinogens, skin
surface area, and soil-to-skin adherence factor. The USEPA has developed
default values for each of these parameters that can be used if no other site- or
chemical-specific data are available. Many state regulatory agencies within the
United States have also made recommendations for their programs. Where
appropriate, default values are available for several potential human receptors
including workers, and adult and child residents.
RBSLs will not be the same for all routes of exposure. This is because the dose
of a contaminant that a receptor receives depends upon the route of exposure
and the concentration of the contaminant in the different environmental media.
For this reason, it is possible to have several RBSLs for a contaminant at a given
site (i.e., one RBSL for each exposure pathway). The management of the site
requires that the lowest of these RBSLs be used to support the overall risk-based
decisions that are made at the site. Considering the nature and composition of
crude oils (i.e., low in volatile or water-soluble components that could partition
into air or water), it is generally the RBSLs for direct contact with hydrocarbonimpacted surface soils that strongly influence E&P site management decisions.
The receptors of concern are most often onsite workers or other non-residential
receptors depending upon the anticipated future land use of the site.
International Risk Programs

Risk-based decision-making programs are not unique to the United States. For
example, Canada, New Zealand, and the Netherlands have all adopted riskbased decision-making into their site management and cleanup programs. Many
other countries are also considering developing their own risk-based decisionmaking programs. This is appropriate because the basic concepts of risk-based
decision-making are equally applicable everywhere in the world. However, in
practice, it is not a simple task. To develop a risk-based decision-making
program that truly protects human health and the environment, it is necessary to
collect information about how people or other receptors may be exposed to
environmental chemicals in the particular country or region of concern. It is not
31
Chapter 1
sufficient to merely assume that decisions made and/or Tier 1 RBSLs derived
for programs developed in other parts of the world are appropriate for a new
application in a new location. Country and/or region-specific information is
necessary because living and working conditions in the Netherlands, Canada, or
the United States may be quite different than in countries like Nigeria or
Thailand. For example, some differences that may affect exposures include the
length of time that people typically live in the same house, whether groundwater
is consumed and how much is consumed, and how many days per year a worker
spends at his job. The Dutch do not include exposure to groundwater in their
risk program because their citizens do not drink groundwater. In the United
States, consumption of groundwater is common and this pathway is an integral
part of risk evaluation.
A brief summary of some of the international risk-based decision-making
programs and Tier 1 RBSLs used in these programs is provided below.
Dutch National Institute of Public Health and the Environment (RIVM)
In 1994, the Dutch RIVM (National Institute of Public Health and the
Environment) published soil and groundwater Target and Intervention Values
[14]. Intervention Values for soil are defined as the concentrations above which
soils are considered to be seriously contaminated, and are applicable to all sites
exceeding 25 m3 in size. Target Values indicate the soil quality levels ultimately
aimed for and are usually based on background chemical concentrations in soil
in the Netherlands. In applying these values to individual sites, it is generally
concluded that further site investigation is required if the soil concentration of a
given chemical exc eeds this criterion:
Intervention Value + Target Value
2
The target values are derived based on human health toxicity information for
each chemical of concern and residential exposure estimates are derived using
the Dutch CSOIL model. Several different residential exposure pathways are
considered including ingestion of soil, ingestion of crops grown in impacted
soil, and inhalation of indoor air. In selecting the overall exposure pathway of
concern, RIVM derives serious soil contamination concentrations (SCCs) for
each of the exposure pathways and then selects the most conservative (i.e., the
lowest) concentration to propose as the overall Intervention Value for that
chemical. A tiered risk approach is not used and commercial exposure scenarios
are not considered in the Netherlands.
Canadian Council of Ministers of the Environment (CCME)
In 1991, CCME published Interim Canadian Environmental Quality Criteria for
Contaminated Sites [15] and, in 1996, released a follow-up report, A Protocol
32
Chapter 1
for the Derivation of Environmental and Human Health Soil Quality Guidelines
[16], to derive Soil Quality Guidelines (SQGs) that will replace the 1991
criteria. The Canadian protocol considers potential exposures to both human
and ecological receptors for given land uses. SQGs are derived based on
exposure scenarios for agricultural, residential/parkland, commercial, and
industrial land uses for jurisdictions in Canada. For each of the four land uses,
CCME selects final generic SQGs for chemicals in soil based on the lowest
value generated by either of the environmental or human health approaches.
The protocol also identifies indirect routes of exposure including consumption
of meat, milk, and produce from agriculture land, and consumption of
homegrown produce from residential land.
In December 2000, the CCME published Canada-Wide Standards for Petroleum
Hydrocarbons (PHCs) in Soil: Scientific Rationale [17] to address different
types of hydrocarbon fractions. Following the approach described by the United
States Total Petroleum Hydrocarbon Criteria Working Group [18], the CCME
has defined four broad hydrocarbon fractions (F) including F1: C6 to C10; F2:
>C10 to C16; F3: >C16 to C34; and F4: C 34+. Aliphatic and aromatic sub-fractions
are handled separately. With respect to management of PHCs, additional factors
considered in CCMEs Tier 1 level include: ignition hazard, odor and
appearance, effects on buried infrastructure, formation of non-aqueous phase
liquids, and socioeconomics and technological capabilities.
New Zealand Ministry for the Environment
In 1999, New Zealands Ministry for the Environment published Guidelines for
Assessing and Managing Petroleum Contaminated Sites in New Zealand [19].
The Ministry presents a tiered risk-based approach to site assessment and
development of soil and groundwater acceptance criteria for petroleum
hydrocarbon contaminated sites. Tier 1 acceptance criteria are summarized in
look-up tables and site-specific criteria used to develop Tier 2 and 3 criteria are
outlined. Tier 1 acceptance criteria have been developed for a wide range of
possible site characteristics rather than a single generic scenario. For example,
the Tier 1 criteria include consideration of various land uses, such as
agricultural, residential, commercial/industrial, and maintenance workers . Also
considered are soil type (8 soil profiles are represented), depth of soil
contamination, e.g., <1 meter, 1-4 meters, and >4 meters, and depth of
groundwater from ground surface, such as 2-4 meters, 4-8 meters, and >8
meters. New Zealand specifies specific compounds of concern for each type of
refined product for use in deriving Tier 1 soil acceptance criteria.
References
1.
United States Environmental Protection Agency, (USEPA), National Oil

and Hazardous Substances Pollution Contingency Plan (NCP), Federal
Register, Vol. 55, No. 46, Subpart A (1990).
33
Chapter 1
2.
United States Environmental Protection Agency (USEPA), Risk

Assessment Guidance for Superfund, Vol. I Human Health Evaluation
Manual, Office of Emergency and Remedial Response, OSWER Directive
9285.7-01a, September 29, 1989.
3.
American Society for Testing and Materials (ASTM), Standard Guide for
Risk-Based Corrective Action Applied at Petroleum Release Sites, ASTM
E1739-95, November 1995.
4.
Risk-Based Corrective Action, ASTM E2081-00, April 2000.
5.
Massachusetts Department of Environmental Protection, Massachusetts

Contingency Plan, 310 CMR 40.0900, October 1, 1993.
6.
Illinois Environmental Protection Agency, Tiered Approach to Corrective

Action Objectives, Illinois Bureau of Land Administrative Code Title 35,
Subtitle G, Chapter I, Subchapter f , January 6, 2001.
7.
Louisiana Department of Environmental Quality, Risk Evaluation/

Corrective Action Program, June 20, 2000.
8.
Texas Natural Resource Conservation Commission, Texas Risk Reduction

Program, August 12, 1999.
9.
United States Environmental Protection Agency (USEPA), Soil Screening

Guidance: Technical Background Document, Office of Emergency and
Remedial Response, EPA/540/R-95/128, Washington, D.C. (1996).
10. United States Environmental Protection Agency, Region IX, Region IX

PRGs Table 2000 Update, San Francisco, California, November 2000.
11. United States Environmental Protection Agency, Region III, Risk-Based
Concentration Table, Philadelphia, Pennsylvania, October 2000.
12. United States Environmental Protection Agency, Region VI, Human
Health Medium-Specific Screening Levels, Dallas, Texas, September
2001.
13. United States Environmental Protection Agency, Integrated Risk
Information
System
(IRIS),
Online
Database
at
www.epa.gov/ncea/iris.htm, National Center for Environmental
Assessment, Cincinnati, Ohio (2001).
14. National Institute of Public Health and Environmental Protection (RIVM),
Human Exposure to Soil Contamination: A Qualitative and Quantitative
Analysis Towards Proposals for Human Toxicological Intervention
Values, Report No. 725201011, Bilthoven, The Netherlands, January
1994.
34
Chapter 1
15. Canadian Council of Ministers of the Environment (CCME), Interim

Canadian Environmental Quality Criteria for Contaminated Sites, CCME
EPC-CS34 (1991).
16. Canadian Council of Ministers of the Environment (CCME), A Protocol
for the Derivation of Environmental and Human Health Soil Quality
Guidelines, CCME-EPC-101E (1996).
17. Canadian Council of Ministers of the Environment (CCME), Canada-Wide
Standards for Petroleum Hydrocarbons (PHCs) in Soil: Scientific Rationale,
Supporting Technical Document, December 2000.
18. Total Petroleum Hydrocarbon Criteria Working Group (TPHCWG),
Analysis of Petroleum Hydrocarbons in Environmental Media, Total
Petroleum Hydrocarbon Criteria Working Group Series, Vol. 1, Amherst
Scientific Publishers, Amherst, Massachusetts, March 1998.
19. Ministry for the Environment, Guidelines for Assessing and Managing
Petroleum Contaminated Sites in New Zealand, Wellington, New Zealand,
ISBN 0478 090641, August 1999.
35
Chapter 2
Technical Basis for Current Soil Management

Levels of Total Petroleum Hydrocarbons
Wayne A. Hamilton,1 H. James Sewell,1 and George Deeley 2
1
Shell Exploration & Production Company
Equilon Enterprises LLC
Introduction
This chapter reviews the technical basis for the 1% total petroleum hydrocarbon
(TPH) management level for land management of exploration and production
(E&P) wastes as recommended by the American Petroleum Institute (API) and
provided in state regulations within the United States (e.g., Louisiana (LA) 29b
and Texas (TX) Rule 91) and practice. The limitations of the 1% TPH
management level are discussed and the need to develop risk-based management
levels based on site-specific information is introduced.
Crude oil from a producing formation and diesel added to drilling mud are the
chief sources of petroleum hydrocarbons in E&P wastes [1]. Gross analysis of
petroleum hydrocarbons within a waste may be reported as oil and grease
(O&G), TPH, or total organic carbon (TOC). O&G is determined by
gravimetric analysis of a solvent extract, TPH is determined by infrared analysis
of a solvent extract, and TOC is analyzed by wet oxidation of a sample. O&G
was the predominant analytical method used to measure petroleum hydrocarbon
content of E&P wastes prior to the 1990s, with TPH becoming the predominant
analytical method since that time.
Initially, a 1% TPH concentration in soil was used as a guidance value for E&P
wastes that were land-managed. The 1% TPH guidance value was developed
based on two criteria: the first was the impact of hydrocarbons on plant life and
the second was hydrocarbon mobility to groundwater. Over time, certain United
States regulatory agencies adopted the 1% TPH guidance value as a regulatory
limit for the hydrocarbon content of land-disposed E&P wastes. At sites where
oily wastes having TPH concentrations greater than 1% have been landdisposed, the regulatory limit of 1% is sometimes used as a soil cleanup
standard. In the paragraphs that follow, the terms guidance value, regulatory
limit, and cleanup standard are used where appropriate; the term
management level is used in instances where any of these terms may be
appropriate.
36
Chapter 2
Development of State E&P Waste

Regulations Within the United States
Prior to the early 1980s, drilling wastes were handled in a manner that met the
requirements of the landowner, that were protective of surface water, and that
met aesthetic requirements. Closure of a site often consisted of drying and
burying the waste materials, and contouring the land surface. The landowner
was either paid for damages incurred to the land surface, or the area was
reclaimed [2]. Major impacts to soils and crops were usually attributed to
excess salts [3,4]. Effects of diesel oil on plants were considered less severe and
of shorter duration than salt damage. During this time frame, regulations were
primarily concerned with the quality of effluent reaching surface water. These
water quality effluent requirements were generally met with conventional oilwater separators and other equipment [5]. In remo te production areas, aesthetic
requirements were met when the landowner was satisfied that the site was
returned to a condition that was consistent with its original use.
In 1976, the Resource Conservation and Recovery Act (RCRA) was enacted to
regulate solid waste and resource recovery in the United States for all but a few
exempt wastes. The wastes that were (and still are) exempt include those
associated with the exploration, development, and production of crude oil or
natural gas [6]. The exemption broadly states that oil and gas wastes should not
be regulated as hazardous waste, if they are being managed under existing
regulatory programs in a way that adequately mitigates or prevents harm to the
environment. Therefore, it has been up to state regulatory agencies to determine
adequate management requirements for oilfield wastes, including oil-containing
wastes.
The State of Louisiana was at the forefront of establishing regulations which
specifically defined environmental standards for E&P wastes [7]. Early
regulations required the protection of surface water from oilfield wastes, though
oilfield wastes were not defined [8]. It was the Louisiana regulation of 1986
that clearly defined non-hazardous oilfield waste (NOW) and set forth
regulatory requirements for O&G in soil, along with other components [9].
Landfarming of drilling pit contents required a final O&G content of the
waste/soil mixture to be less than or equal to 1% dry weight; burial of the
mixture onsite required a final O&G concentration of less than or equal to 3%
dry weight [9].
A review of current upstream TPH regulations in North America (see Table 1)
indicates that they are highly variable, ranging from 100 to 10,000 mg/kg soil.
A few states (e.g., Texas, Colorado, Wyoming, and New Mexico) incorporate
some broad risk concepts for determining appropriate TPH regulatory limits by
evaluating site-specific conditions such as depth to groundwater and proximity
to residential areas.
37
Chapter 2
Table 1. Summary of TPH regulations for E&P sites in 1999 [10].

State/Province
Colorado
Louisiana
Michigan
New Mexico
Texas
Wyoming
Alberta
TPH Regulatory Limit

(mg/kg Soil)
10,000
1,000
10,000
10,000
100; 1,000; 5,000
10,000
1,000 to 10,000
1,000
Comments
Non-Sensitive
Sensitive
Land Treatment of NOW
Was 250 mg/kg
Site Dependent
Railroad Commission Rule 91
Site-by-Site Basis
NOW = Non-Hazardous Oilfield Waste

The State of Colorado has established different TPH limits for sensitive sites
(1,000 mg/kg) and non-sensitive sites (10,000 mg/kg). They define sensitive
areas as those areas vulnerable to potential significant groundwater impacts and
areas subject to concentrated human or wildlife use, such as parks, recreation
sites, urban or suburban areas, and wildlife refuges [11,12]. New Mexico uses
ranking criteria based on the general site characteristics to determine their
relative threat to human health, fresh waters and the environment. The ranking
criteria include depth to groundwater, distance from oil wellhead to water
sources, and distance to surface water body [13].
Only one state, Michigan, has changed its TPH regulations for upstream sites
based upon risk of crude oils to human health. In 1997, the Michigan
Department of Environmental Quality changed the regulatory limit for TPH in
soil at upstream sites from 250 mg/kg to 10,000 mg/kg based upon a risk
evaluation of the crude oils produced in their state [14].
The Technical Basis for the 1% TPH Management Level

Petroleum Hydrocarbon Impact on Soils & Plants
A review of the literature that was used to develop APIs 1% TPH guidance
level indicated that at 1% or less of mixed hydrocarbon, little or no (plant)
yield reduction is expected based on existing information [15]. Also, where the
loading was between 1 and 5% petroleum hydrocarbon in soil, the site recovered
after one growing season [15]. This recovery was the result of hydrocarbon
assimilation by the soil, which is a combination of biodegradation,
evaporative loss, and the binding of the petroleum hydrocarbons to soil organic
matter.
38
Chapter 2
There have been a number of explanations for petroleum hydrocarbon impacts

on soil and plants. Early work made the distinction between acute and chronic
plant injury, and biophysical and biochemical effects [16]. Petroleum
hydrocarbons could exert a direct toxic effect by dissolving plant tissue. Poor
growth was attributed to suffocation of the plants caused by the displacement of
air in the soil pores by oil or the exhaustion of oxygen by increased microbial
activity [16].
Also, there may be interference with plant-soil-water
relationships, and toxicity from sulfides and excess available manganese
produced during the biological decomposition of the hydrocarbons [17].
Damage to cell membranes, reduced transpiration rate, increase in respiration
rate, and inhibited translocation were also implicated [18]. The severity of the
effects noted depends upon the constituents and amount of oil, on the
environmental conditions, and on the species of plant [18]. A distinction has
also been made between rapid or acute injury caused by light oils, and slow or
chronic injury resulting from heavy oils [19].
The following excerpts capture much of the essence of the early work in this
field [20]:
The damage that oil does is due mostly to the prevention of the plant
from obtaining sufficient moisture and air and from ramifying its roots:
very little is due to toxicity, as such.
Crude petroleums are converted to soil organic matter by bacteria and

fungi.
the organic matter improves soil physical conditions.
Thus, the development of the 1% TPH guidance level in soil was originally
based on the issue of toxicity to plants. If the amount of petroleum hydrocarbon
in soil is kept at or below this level, and there are acceptable salt and pH levels,
there should not be significant reductions in plant yield, thus meeting landowner
requirements.
More recent studies confirm the earlier studies on plant growth and germination
for heavy or weathered crude oil. However, light or fresher crude oil may be
found to require more stringent guidance. Chaineau et al. [21] found that the
phytoavailability of complex mixtures of hydrocarbons that have low octanolwater partition coefficients (Kow) is negligible, even when the soil petroleum
hydrocarbon content is as high as 1%. Phytotoxicity was found to be greater for
low molecular weight and aromatic hydrocarbons and varied greatly with
hydrocarbon concentration in the soil and plant species [21]. Salanitro et al.
[22] found a similar molecular weight relationship. Residual TPH in which
germination was not affected (<4 to 27% reduction) in bioremediated soils (10
months treatment) varied from 7,000 to 10,000, from 8,200 to 8,600, and from
39
Chapter 2
1,000 to 1,200 mg/kg for the heavy, medium, and light oily soils, respectively
[22].
Current understanding of toxicity issues suggests that soil toxicity
considerations may expand beyond plant toxicity alone. Other potential
ecotoxicity issues are now being examined. For example, recent studies have
examined earthworms [23] as well as other soil invertebrates [24].
Petroleum Hydrocarbon Impact on Groundwater Resources
If a sufficient quantity of hydrocarbon is released to a soil, the hydrocarbon will
migrate through the soil as a non-aqueous phase liquid (NAPL). As the NAPL
moves, some will remain in the pore spaces of the soil, until there is no longer
sufficient volume of NAPL remaining for migration to occur. At this point, the
NAPL in the soil is said to be at residual saturation. The 1% soil TPH guidance
value was selected to be below the minimum level required for hydrocarbon
mobility (or less than the residual saturation). Therefore, the guidance value of
1% TPH in soil prevents movement of NAPL toward groundwater receptors.
Researchers have confirmed that hydrocarbon migration is not a problem at low
percentages of hydrocarbons in soil. Raymond et al. [25] found that by adding
approximately 2% oil to the top 15 centimeters of soil, 99% remained within the
top 20 centimeters after 1 year. When hydrocarbon loading rates of 3 to 13%
per year were added, no significant oil migration was found below the zone of
incorporation [25]. Brost and DeVaull [26] tried to determine a conservative
NAPL concentration in unsaturated soil below which NAPL would be
immobile. Unsaturated soil samples were saturated with hydrocarbons and then
allowed to drain. The amount of residual hydrocarbon remaining in the soil
pores was then quantified. Brost and DeVaull determined the residual saturation
for middle distillates and fuel oils to range from 0.8 to 5.0% [26]. The
variability is believed to be attributable to experimental method variability,
variation in soil type, unique chemical properties and measurement differences.
When oil enters the soil as a NAPL there is natural separation of the
hydrocarbon constituents due to exposure of the NAPL to the solid phase, vapor
phase, and water phase within the soil. The higher molecular weight compounds
are generally less mobile and stay near the source location, while the lighter
weight compounds migrate deeper into the subsurface because of greater
aqueous solubility [20]. Biodegradation of these compounds may play a role in
preventing transport to groundwater. Webster and Loehr [27] studied the rate of
hydrocarbon release for six soils containing petroleum hydrocarbons. Their
work indicated soil hydrocarbons containing slightly to moderately weathered
diesel range organics were more available for release when compared to
weathered crude oil [27].
40
Chapter 2
A mathematical model (called VADSAT) was developed by API to characterize

the leaching of hydrocarbons from land-disposed wastes [28]. The VADSAT
model was used to predict the fate and transport of selected organic components
contained in E&P associated wastes. The model simulations considered various
input scenarios that included: a) a variety of hydrogeological characteristics, b)
four chemicals [benzene, toluene, ethylbenzene, and xylene (total)], c) six waste
types, d) different disposal scenarios, e) infiltration and biodegradation, and f)
500 and 1,500 feet groundwater well receptor locations. The percent oil content
varied with the different waste type. The waste management scenarios assumed
average oil content from 1% for land spreading/burial to 2.5% for road
spreading.
There were 1,144 VADSAT computations for the various
hydrogeological and waste scenarios described above.
All VADSAT
simulations resulted in groundwater concentrations at the receptor locations that
were less than the United States Environmental Protection Agency (USEPA)
groundwater Maximum Contaminant Levels (MCLs) for benzene, toluene,
ethylbenzene, and xylene (total) in drinking water. Based on the available
studies, the 1% TPH soil management level appears to be justifiable because it is
protective of groundwater resources.
Impact of Biodegradation on Hydrocarbon Concentrations

Biodegradation of petroleum hydrocarbons has been well documented in the
literature and bioremediation is recognized as a cost-effective method to treat
soils and other E&P wastes containing petroleum hydrocarbons. In many
situations, a 1% TPH management level is achievable through biodegradation.
Biodegradation of petroleum hydrocarbons involves the metabolism of certain
hydrocarbon compounds (alkanes, alkenes, aromatics, and polars) by indigenous
soil microbes. The petroleum hydrocarbons are converted to carbon dioxide,
water, and biomass. Many factors have been identified that affect both the
kinetics and the extent of hydrocarbon biodegradation. These include soil
properties such as pH, temperature, moisture, aeration, and nutrient status; as
well as hydrocarbon characteristics.
Studies and reviews in the literature have documented the initial petroleum
hydrocarbon loading rates in soil and the extent of soil hydrocarbon
biodegradation. An industry review prepared for API in 1983 of land treatment
practices indicated that 70 to 90% of oily sludge hydrocarbons that were applied
to surface soils at loading rates of 1 to 5% were removed, primarily through
biodegradation [29]. Loehr et al. [30] studied the treatability of an oily sludge in
field plots in a silty loam soil and demonstrated that 60 to 70% of the initial oil
and grease (2 to 5.5%) hydrocarbons were biodegraded within 2 to 3 years.
Studies have demonstrated that degradative processes in soils attenuate the more
mobile, light-end aromatic and water-soluble petroleum hydrocarbons, leaving
behind the more recalcitrant hydrocarbons with little potential for contaminant
migration. Huesemann and Moore [31] showed that 93% of the saturate and
41
Chapter 2
79% of the aromatic hydrocarbon compounds having carbon numbers in the

range of C10-C44+ were degraded in a sandy soil containing weathered Michigan
crude oil (medium API gravity) with an initial concentration of 3% TPH. The
study also indicated that the polar fraction was resistant to microbial metabolism
and did not degrade during the 5.5-month long test.
Work by Huesemann [32] on the limits and extent of TPH remediation in
different soils showed that 90% of the alkanes and monocyclic saturates and 50
to 70% of the aromatic compounds (<C44) were degraded. Other research
indicates that overall bioremediation effectiveness was dependent upon
hydrocarbon types present and was not affected as much by soil type, nutrient
addition, microbial populations, or treatment conditions [22]. Recently, a study
of bioremediation showed that after bioremediation, petroleum hydrocarbons in
oily soil decreased from 70 to 90%, from 40 to 60% and from 35 to 60% for
those carbon number species in the range of C11C22, C23C32, and C35C44,
respectively [22]. In a pilot study designed to determine the fate of hydrocarbon
constituents during land treatment of soil impacted with fresh Michigan crude
oil (3.1 wt % TPH), results indicated that biodegradation was the primary
pathway for TPH removal, accounting for 94% removed in 1 year [33].
The 1% TPH soil management level seems to be supported by both earlier and
more recent investigations of petroleum hydrocarbon biodegradation. Cleanup
standards that are less than 1%, while achievable in some situations, would be
difficult to achieve at sites containing residual weathered petroleum
hydrocarbons.
Summary
The 1% TPH management level used by Texas, Louisiana, and other regulators
has been shown to be protective of groundwater resources and plant life. Also,
the 1% TPH management level was shown to be achievable, specifically through
bioremediation.
Although the 1% TPH management level may be adequate in many cases, riskbased cleanup standards are still needed to address the human health risks of
particular hydrocarbon mixtures. The specific components of crude oil must be
identified and the potential risks evaluated for those sites at which human
exposures may be of concern. For this reason, within the last 10 years, technical
methods have been developed to determine soil and groundwater corrective
action cleanup criteria using a site-specific and human health risk-based
approach. These methods are summarized in this book, and crude oil
composition data are provided to aid in the risk evaluation. Risks to ecological
receptors (other than plants) may also need to be considered, but the state of the
science is not as advanced as that for human health risk assessment.
42
Chapter 2
References
1.
Deuel, Jr., L.E., and Holliday, G.H., Soil Remediation for the Petroleum
Extraction In dustry, PennWell Books, Tulsa, Oklahoma (1994).
2.
Allred, R.B., Drilling Operations On Land, proceedings of a

symposium: Research on Environmental Fate and Effects of Drilling Fluids
and Cuttings, Lake Buena Vista, Florida, January 21-24, 1980.
3.
Miller, R.W., and Honarvar, S., Effect of Drilling Fluid Components and
Mixtures on Plants and Soils, conference proceedings: The Environmental
Aspects of Chemical Use in Well Drilling Operations, Environmental
Protection Agency, EPA/1-75-004 (1975).
4.
Pesaran, P., Effect of Drilling Fluid Components and Mixtures on Plants

and Soils, MS Thesis, Utah State University (1977).
5.
McGlothlin, R.E., and Krause, H., Water Base Drilling Fluids,

proceedings of a symposium: Research on Environmental Fate and Effects
of Drilling Fluids and Cuttings, Lake Buena Vista, Florida, January 21-24,
1980.
6.
Burford, R., and Flannery, D.M., Developing Regulatory Concepts

Impacting on the Management of Wastes Produced by Oil and Gas Well
Drilling Operations, proceedings of a National Conference on Drilling
Muds, University of Oklahoma, Norman, May 28-29, 1987.
7.
Wascom, C.D., Oilfield Pit Regulations, A First for the Louisiana Oil &
Gas Industry, proceedings of a National Conference on Drilling Muds,
University of Oklahoma, Norman, May 29-30, 1986.
8.
Louisiana Department of Natural Resources, Statewide Order 29-B, Drilling

and Production, Chapter XV, Office of Conservation (1967).
9.
Louisiana Department of Natural Resources, Statewide Order 29-B, Drilling

and Production, Chapter XV, Office of Conservation (1986).
10. McMillen, S.J., Magaw, R. I., Kerr, J. M., Sweeney, R. E., Nakles, D. V.,
and Geiger, S.C., A New Risk-Based Approach to Establish Cleanup
Levels for Total Petroleum Hydrocarbons, Proceedings of the 6th
International Petroleum Environmental Conference, Houston, Texas,
pp. 438-459, (2000).
11. Colorado BLM Policy for the On-Site Remediation of RCRA Exempt,
Hydrocarbon Based, Oil and Gas Exploration & Production Waste and
Terrestrial Spills of Crude Oil, June 1997.
12. Colorado Oil & Gas Co mmissions (COGCC) Rules 907 and 908 (1995),
and Revisions to the 900 Series Rule (1997).
43
Chapter 2
13. New Mexico Oil Conservation Division (OCD), Guidelines for

Remediation of Leaks, Spills, and Releases, August 1993.
14. Michigan Department of Environmental Quality, Supervisors Letter No.
1997-1, Total Petroleum Hydrocarbon (TPH) and Aesthetic Cleanup
Criteria for Soil at Oil and Gas Facilities, May 1997.
15. Deuel, Jr., L.E., Evaluation of Limiting Constituents Suggested for Land
Disposal of Exploration and Production Wastes, American Petroleum
Institute Publication No. 4527, July 1991.
16. Currier, H.B., and Peoples, S.A., Phytotoxicity of Hydrocarbons,
Hilgardia, 23, 155-173, November 1954.
17. Udo, E.J., and Fayemi, A.A.A., The Effect of Oil Pollution of Soil on
Germination, Growth, and Nutrient Uptake of Corn, J. Environ. Qual., 4,
537-540 (1975).
18. Baker, J.M., The Effects of Oils on Plants, Environ. Pollut., 1, 27-44
(1970).
19. deOng, E.R., Knight, H., and Chamberlin, J.C., A Preliminary Study of
Petroleum Oil as an Insecticide for Citrus Trees, Hilgardia, 2, 353-384
(1927).
20. Plice, M.J., Some Effects of Crude Petroleum on Soil Fertility, Soil
Science Society Proceedings 1948, 13, 413-416 (1948).
21. Chaineau, C.H., Morel, J.L., and Oudot, J., Phytotoxicity and Plant Uptake
of Fuel Oil Hydrocarbons, J. Environ. Qual., 26, 1478-1483 (1997).
22. Salanitro, J.P., Dorn, P.B., Huesemann, M.H., Moore, K.O., Rhodes, I.A.,
Rice Jackson, L.A., Vipond, T.E., Western, M.M., and Wisniewski, H.L.,
Crude Oil Hydrocarbon Bioremediation and Soil Ecotoxicity Assessment,
Environmental Science & Technology, 31, 1769-1776 (1997).
23. Saterbak, A., Toy, R.J., Wong, D.C.L., McMain, B.J., Williams, M.P.,
Dorn, P.B., Brzuzy, L.P., Chai, E.Y., and Salanitro, J.P., Ecotoxicological
and Analytical Assessment of Hydrocarbon-Contaminated Soils and
Application to Ecological Risk Assessment, Environmental Toxicology
and Chemistry , 18, 1591-1607 (1999).
Standards for Petroleum Hydrocarbons (PHCs) in Soil:
Scientific
Rationale, Supporting Technical Document, December 2000.
25. Raymond, R.L., Hudson, J.O., and Jamison, V.W., Oil Degradation in
Soil, Applied and Environmental Microbiology, 31, 522-535 (1976).
26. Brost, E.J., and DeVaull, G.E., Non-Aqueous Phase Liquid (NAPL) FreePhase Mobility Limits in Soil, Appendix 5, Conservative Estimates of
44
Chapter 2
Residual Phase Mobility, In Draft: Development of Tier 1 Risk-Based

Corrective Action (RBCA) Tools for Application at Exploration and
Production Facilities, Gas Research Institute Contract No. 5097-210-3874
(1999).
27. Webster, M.T., and Loehr, R.C., Evaluation of Chemical Release to
Groundwater from Contaminated Soils, The 5th International Petroleum
Environmental Conference, Albuquerque, New Mexico, October 20-23,
1998.
28. American Petroleum Institute (API), Characterization of Exploration and
Production Associated Waste, Publication Number DR53, November
1996.
29. American Petroleum Institute (API), Land Treatment Practice in the
Petroleum Industry, prepared by Environmental Research & Technology
Inc., June 1983.
30. Loehr, R.C., Martin, Jr., J.H., and Neuhauser, E.F., Land Treatment of an
Aged Oily Sludge-Organic Loss and Change in Soil Characteristics, Water
Resources , 26, 805-815 (1992).
31. Huesemann, M.H., and Moore, K.O., The Effects of Soil Type, Crude Oil
Type and Loading, Oxygen, and Commercial Bacteria on Crude Oil
Bioremediation Kinetics as Measure by Soil Respirometry, Journal of Soil
Contamination, 2, 245-264 (1993).
32. Huesemann, M. H., Predictive Model for Estimating the Extent of
Petroleum Hydrocarbon Biodegradation in Contaminated Soils,
Environmental Science and Technology, 29 (1), 7-18 (1995).
33. Saterbak, A., Rouse, S.J., and Caico, C.A., Fate of Hydrocarbon
Constituents During Land Treatment, Internal Shell Western E&P
Document (1997).
45
Chapter 3
Composition of Crude Oils and Gas

Condensates
Sara J. McMillen,1 Jill M. Kerr,2 and David V. Nakles 3
1

ExxonMobil Production Company
3
2
Introduction
One of the basic principles of risk-based decision-making is that the decisionmaker must have a good understanding of the composition of the material of
concern. Therefore, knowledge of the chemical and physical characteristics of
crude oils and condensates is required for the effective application of risk-based
decision-making at exploration and production sites. A summary of the
chemical and physical properties of both condensates and crude oils is presented
in this chapter. Crude oils are defined as mixtures of hydrocarbons that exist in
the liquid phase in underground reservoirs and that remain in the liquid phase at
atmospheric pressure. Condensates are mixtures of hydrocarbons that are in the
vapor phase under reservoir pressures and temperatures but become liquid under
atmospheric conditions.
Molecular Structure of Hydrocarbons

Petroleum hydrocarbons are organic compounds comprised of carbon and
hydrogen atoms arranged in varying structural configurations. In the broadest
sense, petroleum hydrocarbons can be divided into two classes of chemicals, the
saturates, which have only single bonds between carbon atoms, and the
unsaturates, which have at least one double bond between carbon atoms
(Figure 1). The saturates, also referred to as alkanes or paraffins, are comprised
of three main subclasses of compounds based on the structure of their
molecules, either straight chains, branched chains, or cyclic (see Figure 2). (The
terms saturated and aliphatic hydrocarbons are interchangeable and both are
used to describe this group of compounds.) Straight-chain compounds are
known as normal alkanes (or n-alkanes). The branched-chain compounds are
designated isoalkanes and the cyclic, or ring-like compounds, cyclo alkanes.
More familiar terms used by petroleum geologists to describe these structures
are paraffins for alkanes and naphthenes for cycloparaffins or cycloalkanes.
Within the unsaturates, there are two main subclasses of compounds, aromatics
and olefins. Aromatic hydrocarbons are comprised of one or more unsaturated
cyclic structures, or rings. Benzene contains one such ring, while polycyclic
46
Chapter 3
aromatic hydrocarbons (PAHs) contain two or more rings (e.g., phenanthrene

has three unsaturated rings). Olefins contain double bonds between two or more
carbon atoms. Olefins are not found in crude oil or condensates because they
are readily reduced with hydrogen to paraffins in the reservoir [1]. Olefins are
formed during refining of crude oils and they are present in most refined
products. A classification of petro leum hydrocarbons by structure or molecular
type is shown in Figures 1 and 2.
Saturates:
(alkanes or paraffins)
Unsaturates:
Olefins:
n-Alkane:
H
H
H
H | H | H | H
| C | C | C |
C | C | C | C
| H | H | H |
H
H
H
H
H
|
C
H
H
H
H | H | H | H
| C | C | C |
C | C | C | C
| H | H | H |
H
H
H
H
|
C
|
H H H
Aromatics:
Isoalkane:
H
H
C = C
C
|
C
H
H
C
||
C
C
|
H
Double
Carbon
Bond
H
Cycloalkane:
H
H
H
H
C
C
C
C
C
H H
Figure 1. Examples of petroleum hydrocarbon structures.
47
Chapter 3
Petroleum
Hydrocarbons
Saturates (also
known as paraffins
or alkanes)
n-alkanes
(straight
chain)
isoalkanes
(branched
chain)
Unsaturates
cycloalkanes
or naphthenes
(cyclic)
Aromatics
Olefins
Figure 2. Chemical classification of petroleum hydrocarbons [2].
Crude Oil Composition

Crude oil is composed almost entirely of the elements hydrogen and carbon, in
the ratio of approximately 1.85 (hydrogen):1 (carbon). In addition to the
hydrocarbons, there are also two non-hydrocarbon fractions that contain
elements in addition to carbon and hydrogen, such as nitrogen, sulfur, and
oxygen. These elements constitute less than 1% to as much as 7% of some
crude oils [1]. These non-hydrocarbon fractions are the asphaltenes and resins.
Hydrocarbons comprise the majority of the components in most crude oils and
are the compounds that are primarily, but not always, measured as total
petroleum hydrocarbons (TPH). The primary saturated and unsaturated
hydro carbons consist of n-alkanes, isoalkanes, cycloalkanes, and the mono-, di-,
and tri-aromatics; there are no olefins in crude oil. Crude oils vary in
appearance from straw yellow, green, and brown to dark brown or black in color
[3].
Crude Oil Classification
Petroleum geologists often classify crude oils based on their hydrocarbon class
composition. Several classification schemes have been published [4,5,6],
including that shown in Figure 3. The composition of an example crude oil is
presented in Table 1. This oil is a 35API gravity oil that would be classified as
a naphthenic oil.
48
Chapter 3
Paraffinic
Pennsylvania
Rumania
Poland
Mid-continent
Baku
Naphthenic
Grozny
Borneo
Borneo
Japan
Aromatic
Mexico
Rumania
N. Calif
Gulf
Coast
Asphaltic
Figure 3. Classification of crude oils based on relative amounts of paraffins,

naphthenes, asphaltenes, and aromatics [6].
Table 1. Composition of a naphthenic 35API-gravity crude oil [1].
Molecular Type
Paraffins
Naphthenes
Aromatics
Asphaltenes
Total
Weight Percent
25
50
17
8
100
The composition of 636 crude oils from around the world have been compared
by Tissot and Welte as shown in Figure 4 [7]. These data reveal that the
proportions of saturates, aromatics, resins, and asphaltenes can vary dramatically
among crude oils , with the majority of crude oils lying within a composition
envelope that is bounded in the following manner:
40 to 80% Saturates
15 to 40% Aromatics
0 to 20% Resins and Asphaltenes
Tissot and Welte found that 95% of the crude oils produced around the world
fell into this distribution pattern as shown in Figure 4.
49
Chapter 3
Aromatics 100%
95% Frequency
Distribution Envelope
World Oils
50%
100%
Saturates
50%
50%
100%
Resin + Asphaltenes
Figure 4. Ternary diagram showing the class composition of crude oils.

Carbon Number
It would be extremely difficult to identify all of the components of crude oils
and fuels, so petroleum and petroleum products are characterized in terms of
boiling range and approximate carbon number. Petroleum products can be
classified by their distillation temperature, or boiling point ranges, which is also
an indication of the carbon number range of each fuel. Figure 5 shows boiling
points and carbon ranges for six common crude oil products [8]. The
composition (in terms of refined petroleum products and carbon ranges) of a
35API-gravity oil is shown in Table 3 [1].
50
Chapter 3
Figure 5. The boiling point ranges and carbon ranges for six common crude oil
products [8].
Table 3. Composition of a 35API-gravity crude oil [1].
Molecular Size
Gasoline (C5-C10)
Kerosene (C8-C12)
Diesel Fuel (C13-C17)
Heavy Gas Oil (C19-C25)
Lubricating Oil (C20-C45)
Residuum (>C40 )
Total
Volume Percent
27
13
12
10
20
18
100
Within each of the different classes of hydrocarbons (saturates or aromatics) are

compounds that have anywhere from 1 to more than 45 carbons in their
chemical structure. The percentages of these compounds that are present vary
among different crude oils. This characteristic can be observed by analyzing
crude oils and fuels by gas chromatography.
Figure 6 illustrates the distinctive gas chromatography fingerprints of gasoline,
diesel, and two crude oils. Gas chromatograms, or fingerprints, give an
indication of the carbon number range and hydrocarbon type (saturates versus
aromatics) for the total petroleum hydrocarbons within a complex mixture. As
shown in Figure 6, the diesel fuel or gasoline signatures contain hydrocarbons in
the appro ximate range of C10-C24, and C6-C12, respectively. Therefore, their
hydrocarbon ranges are always more narrow than those for crude oils. The
51
Chapter 3
Widuri crude oil from Sumatra is dominated by normal alkanes or paraffins that
produce a picket fence type pattern in the chromatograph, which is typical of
waxy crude oils. On the other hand, the San Joaquin Valley (SJV) crude oil
from California is dominated by a hump or unresolved complex mixture of
hydro carbons that is difficult for a gas chromatograph to separate. This hump is
indicative of the prior biodegradation of hydrocarbons that occurred in the oil
reservoir and is a common characteristic of many heavy crude oils.
Gasoline
Widuri Crude Oil
52
Chapter 3
Diesel
SJV Crude Oil
Figure 6. Gas chromatography fingerprints of gasoline, Widuri crude oil, that is

enriched in normal alkanes that appear as a picket fence type signature, a
diesel fuel and a SJV crude oil that has a hump that represents a large
unresolved complex mixture. Units are Intensity (mV) vs. Time (minutes).
Gas Condensate Composition
Gas condensates are extracted with natural gas in a liquid form. They have a
narrower carbon number range than crude oil, typically <C6-C15. However,
53
Chapter 3
many condensates that come directly from wells tend to have a tail of heavier
hydrocarbons, while condensates that come from natural gas processing plants
or from condensation in pipelines have a much narrower range. This is because
some amount of processing has occurred, and these types of condensates will
have carbon ranges similar to refined gasoline.
Gas chromatography fingerprints of the saturated and aromatic hydro carbon
fractions of two condensates are shown in Figure 7. These fingerprints illustrate
the large degree of variability that can exist for these hydrocarbon mixtures. In
particular, it is clear that Condensate A encompasses a much broader range of
hydro carbons than does Condensate B. Also, the ratio of the saturated
hydrocarbons to the aromatic hydrocarbons is quite different for these two
condensates, increasing from 3.2 for Condensate B to 5.8 for Condensate A.
Condensate B
Condensate A
Saturate
Saturate
* * *
Aromatic
Aromatic
*
*
* Internal Standard
Figure 7. Gas chromatograms of condensates.
Compounds of Concern in
Crude Oils and Gas Condensates
Following crude oil or condensate releases to the environment, regulators
usually require that soils and/or groundwater be analyzed for TPH. Some
jurisdictions may also require analysis for other compounds of concern (COCs).
These COCs might include specific volatile hydrocarbons such as benzene,
toluene, ethylbenzene, and xylenes; the 16 semi-volatile polyaromatic
hydrocarbons that are on the EPAs priority pollutant list; and some metals. The
amount of each of these COCs in crude oils and condensates as well as the
implications of their presence on risk-based decision-making for E&P sites are
discussed in detail in Chapters 10 through 12.
54
Chapter 3
An Overview of the Physical

Properties of Petroleum Hydrocarbons
The fate and transport of a hydrocarbon mixt ure in the environment is an
important aspect of risk assessment because it determines the exposure of a
human or ecological receptor to the mixture. The key physical characteristics of
hydrocarbons that affect their fate and transport in the environment include:
Solubility in Water: This property is arguably the most important

factor that determines the transport of hydro carbons in groundwater or
surface water.
Volatility: The volatility of a hydrocarbon will dictate its movement

with air or other gases.
Density: The density of a hydrocarbon is expressed as its API gravity,

which is a measure of its specific gravity. The API gravity is inversely
proportional to the specific gravity of the compound at 60F (15C) and
is expressed as an integer [9]. It has units of degrees. As a point of
reference, fresh water has an API gravity of 10. The API gravity of
refined products varies with the specific product, dropping as low as
15 for No. 6 Fuel Oil and as high as 62 for gasoline.
API Gravity =
141 .5
Specific Gravity @ 60 F
131.5
Viscosity: This parameter is a measure of the internal resistance of a

fluid to flow. Highly viscous material, like molasses, does not flow
easily under the forces of gravity while water, a low viscosity material,
flows readily. The viscosity of a fluid tends to decrease with an
increase in temperature.
Pour Point: The pour point is the temperature below which a crude oil
will not flow in a horizontal tube [1]. The pour point for most oils
arises from the precipitation of wax such that a pasty, plastic mass of
interlocking crystals is formed. Wax-free oils have pour points that are
dependent upon viscosity only and will tend to thicken to glassy
materials as the temperature is reduced and the viscosity increases.
Some waxy crude oils may be solid at temperatures as high as 110F
(43C).
Crude Oil
Crude oil composition varies greatly and the differences in composition are
reflected in the API gravity values for crude oils. For example, heavy API
gravity oils (<20API) have higher concentrations of asphaltenes and resins than
55
Chapter 3
do light API gravity oils (>30API). Crude oils produced in the United States
have API gravity values that range from approximately 7 to 50 [6].
Crude oil is less dense than water with a specific gravity ranging from 0.85 to
0.98 (as compared to 1.0 for water). However, because of the large differences
in composition among the various crude oils, the precise specific gravity of the
crudes can vary substantially.
Crude oil also tends to be a viscous liquid at surface temperatures and pressures.
Surface viscosity values range from 1.9 to 19,400 centistokes [3]. Pour point
values for crude oils range from 70 to 110F [1]. Therefore, some crude oils
may be solid at typical seasonal fall and spring temperatures in the United
States. The viscosity and pour point are important because they imply that many
crude oils are not fluid enough to rapidly percolate through soil.
Crude oil is sparingly soluble in water, with solubility increasing with API
gravity. For example, a crude oil with an API gravity of 11 had a total
solubility in water of 3.5 mg/L at 25C (77F), whereas an oil with an API
gravity of 28 had a solubility of 65 mg/L [10]. However, total solubility is
dependent on temperature and the composition of the crude oil.
Condensates
Extensive physical property data are not currently available for condensates.
However, in broad terms, these hydrocarbon mixtures generally exhibit an API
gravity of greater than 45. This suggests that they are not extremely viscous at
normal ambient temperatures and that they are relatively volatile and soluble in
water. Composition data [11] for four condensates revealed that high molecular
weight alkanes can be present. The presence of these alkanes would have a
tendency to increase both density (i.e., decrease API gravity) and viscosity and
decrease both solubility and volatility of the hydrocarbon mixture.
Summary
The composition of crude oils and condensates can vary greatly as reflected by
their class compositions, carbon ranges , and other properties such as API
gravity. Composition may affect fate and transport in the environment, and
therefore can impact risk-based decision-making. While it is possible to
generalize as to crude oil and condensate content and properties, understanding
the specific composition of oils and condensates is required before
implementing risk-based decision-making at E&P sites. For this reason, the
Petroleum Environmental Research Forum Project 97-08 embarked on analyzing
a large number of crude oils and condensates that would be representative of
those produced around the world. The data from this projectare presented in the
following chapters.
56
Chapter 3
References
1.
Hunt, J.M., Petroleum Geochemistry and Geology, San Francisco, W.H.

Freeman and Company (1979).
2.
API Publication No. 4709, Risk Based Methodologies for Evaluating

Petroleum Hydrocarbon Impacts at Oil and Natural Gas E&P Sites (2001).
3.
Selley, R.C., Elements of Petroleum Geology, W.H. Freeman & Company,

(1985).
4.
Smith, N.A.C., The Interpretation of Crude Oil Analysis, U.S. Bureau of

Mines Report No. 2806 (1927).
5.
Sachenen, A.N., Chemical Constituents of Petroleum, New York,

Rheinhold (1945).
6.
Bruce, L.G., and Schmidt, G.W., Hydrocarbon Fingerprinting for

Application in Forensic Geology: A Review With Case Studies, AAPG
Bulletin, 78 (11), 1692-1710 (1994).
7.
Tissot, B.P., and Welte, D.H., Petroleum Formation and Occurrence, Berlin:
Springer-Verlag (1978).
8.
ASTM, Manual on Hydrocarbon Analysis , 4th Ed., American Society for

Testing and Materials, West Conshohocken, Pennsylvania (1989).
9.
Miles, J.A., ed., Illustrated Glossary of Petroleum Geochemistry, Oxford

University Press (1989).
10. Western States Petroleum Association, Chemical and Physical

Characteristics of Crude Oil, Gasoline, and Diesel Fuel: A Comparative
Study, Glendale, California (1993).
11.
Hawthorne, S.B., and Miller, D.J., Hydrocarbon Characterization of
Gas Condensates and Impacted Soils, Proceedings of the International
Petroleum Environmental Conference, Albuquerque, New Mexico, October 2023, 1998.
57
Chapter 4
Application of the Total Petroleum

Hydrocarbon Criteria Working Group
(TPHCWG) Methodology to Crude Oils and
Gas Condensates
Sara McMillen,1 Ileana Rhodes ,2 David V. Nakles,3 and Robert E. Sweeney4
1

3
4
Unocal Corporation
2
Introduction
The Total Petroleum Hydrocarbon Criteria Working Group (TPHCWG)
convened in 1993 to develop scientifically defensible information for
establishing soil cleanup levels that are protective of human health at petroleum
release sites [1-5]. The impetus for the formation of this group was the large
disparity in cleanup standards for total petroleum hydrocarbons (TPH) in soil,
and the recognition that these regulatory standards were not based on risk to
human health. Active participants in this effort were the Air Force, Exxon,
Shell, Chevron, the American Petroleum Institute (API), the Association of
American Railroads, several state governments (Washington, Texas, Colorado,
Hawaii, Louisiana, New Mexico, Massachusetts) of the United States , the
United States Environmental Protection Agency (USEPA), the Department of
Defense, as well as private consulting companies including EA Engineering
Science and Technology and Menzie-Cura & Associates, Inc.
The use of risk-based decision-making for petroleum mixtures is complicated
due to the fact that petroleum consists of several thousand individual
hydrocarbons and other compounds, each with a unique set of physical and
chemical characteristics including volatility and solubility. Only about 250 of
these compounds have been specifically identified, and it is impossible to
analyze all of the specific constituents in most petroleum products or crude oils.
In response to this difficulty, the TPHCWG chose to use a fractionation
approach to analyze for TPH and coupled this data with the standard risk
assessment approach for deriving Tier 1 risk-based screening levels (RBSLs)
described by the American Society for Testing and Materials (ASTM) [6].
Similar approaches for deriving TPH RBSLs or cleanup levels have been
adopted by the States of Massachusetts, Texas, and Louisiana and incorporated
into their environmental management programs for downstream sites [7-9].
58
Chapter 4
The TPHCWG summarized their findings in five volumes [1-5]. This chapter
provides a brief overview of their major findings, and presents new data on the
composition of crude oils and condensates. Before considering the new
analytical approach developed by the TPHCWG, the limitations of older TPH
analytical methods are described.
Definition of Total Petroleum Hydrocarbons (TPH)

TPH is not specifically a measure of petroleum hydrocarbons, but is rather a
measurement of the compounds that are soluble in certain solvents and detected
by various techniques (infrared, gravimetric, gas chromatography). Many
compounds other than petroleum hydrocarbons (plant waxes, soil humic
material, animal fats, etc.) may be measured as TPH. Furthermore, the same
sample analyzed by different TPH methods will produce different TPH
concentrations due to differences in solvent type, extraction method, detection
method, and quantification standards. TPH is therefore defined by the analytical
method that is used to measure it.
Conventional bulk measurements of TPH in a sample are not sufficient to
support a human health risk assessment. To illustrate this point, high TPH
concentrations can be measured in items that clearly do not pose a risk to human
health. For example, TPH concentrations have been measured in many items
found in nature including peat moss (3,700 mg/kg of TPH), pine needles (19,000
mg/kg of TPH), cow manure (12,000 mg/kg of TPH), and hay (4,500 mg/kg)
[10]. It has also been measured in household petroleum jelly at concentrations
of 749,000 mg/kg [11]. Although these TPH concentrations are substantially
greater than many existing TPH regulatory standards, none of these materials are
considered a risk to human health.
Review of TPH Analytical Methods

Some of the more common methods for the analysis of TPH include: (1) Method
418.1 or Modified 418.1, (2) Method 413.1 for oil and grease, (3) Modified
8015M for Diesel-Range Organics (DRO), and (4) Modified 8015M for
Gasoline-Range Organics (GRO) [12]. Method 418.1 consists of solvent
extraction followed by treatment in a silica gel column and infrared
spectroscopy; the modified Method 8015 for DRO and GRO are solvent
extractions followed by gas chromatography (GC). If it is suspected that the
sample is predominately a gasoline (i.e., volatile) fraction, purge and trap
sample introduction to the gas chromatograph is often used in the determination
of GRO. Method 413.1 is a gravimetric method that consists of solvent
extraction, evaporation of the solvent, and a weight measurement.
In addition to these standard methods, it should be recognized that there are
many permutations of these methods that have been developed and applied by
59
Chapter 4
environmental regulatory agencies as well as by individual commercial

analytical laboratories. These permutations evolved because, historically, no
one universal method for the measurement of petroleum hydrocarbons was
available for use. Many of these methods are modified versions of the gas
chromatographic methods and are referred to as modified 8015.
Figure 1 shows the overlap between the carbon number ranges of different
hydrocarbon products as well as the overlap in the corresponding TPH analytical
methods [1]. For example, this figure demonstrates that a TPH method designed
for gasoline range organics (i.e., C6 to C12) may report some of the hydrocarbons
present in diesel fuel (i.e., C10 to C28). The same is also true for TPH analytical
tests for diesel range organics which will identify some of the hydrocarbons
present in gasoline-contaminated soils. Lastly, TPH Method 418.1 covers the
complete range from gasoline through lube oil, motor oil, and grease (i.e., C8 to
C40). However, crude oils may contain hydrocarbons with carbon numbers that
range from C3 to C45+ and are not fully addressed even with the use of all three
TPH methods.
Gasoline
Diesel Fuel/Middle Distillates
Lube/Motor Oil, Grease
C2 C 4 C6 C8 C 10C1 2 C14 C16 C18 C2 0 C2 2C2 4C26 C28 C30

TPH Methods: Approximate Carbon Ranges
Purgeable/Volatile/Gasoline Range, Modified 8015, Purge and Trap, GC
Diesel Range, Modified 8015, Extraction, GC
418.1, Modified 418.1: Extraction, Infrared
Figure 1. Carbon number ranges addressed by TPH analytical methods [1].

The hazard evaluation that is conducted as part of the risk evaluation of a site
requires some level of understanding of the chemical composition of the
hydrocarbons that are present in the soil and groundwater. The traditional TPH
analytical techniques are not adequate to support this hazard evaluation because
they provide no specific information about the hydrocarbons that are detected.
The TPHCWG Approach to Assessing Risk of TPH

The general approach of the TPHCWG consists of an assessment of risk
associated with both cancer and non-cancer health effects. Some specific
hydrocarbons have been identified as potential carcinogens: benzene and the
60
Chapter 4
seven carcinogenic polyaromatic hydrocarbons (i.e., benz[a]anthracene,

chrysene, dibenz[a,h]anthracene, benzo[a]pyrene, benzo[b]fluoranthene,
benzo[k]fluoranthene, and indeno[1,2,3-cd]pyrene). These compounds are
quantified separately and evaluated using standard methods of risk assessment.
Discussions of cancer health effects as they relate to benzene and polyaromatic
hydrocarbons are presented in Chapters 10 and 11.
Prior to the publications from the TPHCWG there was no standard method for
assessing the potential risk of non-cancer health effects caused by complex
mixtures of hydrocarbons or TPH. It is in this area that major innovations in the
risk assessment methodology have been made. These innovations focused on
the development of a better understanding of the composition of the refined
products of petroleum and assigning toxicity, fate, and transport characteristics
to hydrocarbon fractions.
The complexity of petroleum hydrocarbons represented the major challenge to
the TPHCWG. Table 1 illustrates that this complexity is due to the number of
possible isomers that may be present in higher molecular weight mixtures such
as diesel or crude oils.
Table 1. Possible number of paraffin isomers for each size molecule [13].
Size
C1, C2 , C3
C4
C5
C6
C7
C8
C9
Isomers
1 Each
2
3
5
9
18
35
Size
C10
C11
C12
C13
C15
C18
C25
Isomers
75
159
355
802
4,347
60,523
36,797,588
Because it is impossible to analyze complex petroleum hydrocarbon mixtures

for all of their constituents, and no one compound could possibly act as a
surrogate for these mixtures, the TPHCWG chose to use a fractionation
approach to assess oil composition. The first step in this fractionation approach
was to separate the hydrocarbons into two groups based on chemical structure
(i.e., aliphatic hydrocarbons and aromatic hydrocarbons). Once segregated into
these groups, the aliphatic hydrocarbons were separated into six carbon number
fractions and the aromatic hydrocarbons into seven carbon number fractions
(Figure 2). Each of the 13 fractions was then treated as if it were a separate
compound in the environment.
61
Chapter 4
Aromatics
H
|
C
HC
CH
HC
CH
Aliphatics
H H H
H H H
H H
|
|
|
|
|
|
|
|
H C C C C C C C C H
|
|
|
|
|
|
|
|
H H H
H H H H H
C
|
H C H
|
H
>C6 -C7 (benzene)

>C7 -C8 (toluene)
>C8 -C10
>C1 0-C12
>C1 2-C16
>C1 6-C21
>C2 1-C35
>C5 -C6
>C6 -C8
>C8-C10
>C10-C12
>C12-C16
>C16-C35
Figure 2. TPH Fractionation: Separation of chemical groups into carbonnumber ranges .

The TPHCWG developed an analytical technique that is based on USEPA SW846 [13] methods for separating hydrocarbons into fractions using GC
techniques. First, the petroleum hydrocarbon or pentane extract of a soil is
separated into aliphatic and aromatic hydrocarbon fractions. This chemical
separation is based on an alumina column procedure (SW-846 USEPA Method
3611) or a silica gel column procedure (SW-846 USEPA Method 3630) [13].
The aliphatic and aromatic hydrocarbon fractions are analyzed separately by GC
and quantified by summing the signals within a series of specific carbon ranges.
The GC is equipped with a boiling point (i.e., non-polar capillary) column [1].
The 13 TPH fractions are based on equivalent carbon (EC) numbers rather
than carbon numbers. ECs are related to the boiling point of individual
compounds in a boiling point GC column, normalized to the boiling point of a
normal alkane. Thus, for compounds where only a boiling point is known, the
EC can be readily calculated. For example, the EC of benzene is 6.5 because its
boiling point and GC retention time are approximately halfway between those of
n-hexane and n-heptane. Benzenes EC number is greater than that of n-hexane
because its ring structure results in a higher boiling point. The TPHCWG chose
the concept of EC numbers because these values are more logically related to
compound mobility in the environment than carbon numbers [3].
62
Chapter 4
USEPA Method 8260 is required to analyze for benzene (the only hydrocarbon
in the aromatic C6-C7 fraction), and toluene (the only hydrocarbon in the
aromatic C7-C8 fraction), rather than using the TPHCWG analytical method.
The TPHCWG analytical method is provided in the Appendix.
The Basis for Defining 13 TPH Fractions
The EC fractions were identified by selecting groups of hydrocarbons that have
similar fate and transport properties, such as solubility and vapor pressure. This
was done because of the important role that fate and transport play in
determining the exposure of a receptor to a site contaminant. For example,
highly soluble petroleum compounds are more likely to migrate to groundwater
and represent potential risk to humans via the consumption of drinking water.
By choosing fate and transport criteria for the definition of the fractions, the
TPHCWG ensured that the risk assessment would properly capture the fraction
of the hydrocarbon mixture that would be present at the point of exposure for the
variety of exposure pathway-receptor combinations that might be present at a
site.
Aliphatic and aromatic hydrocarbon fractions were considered separately
because their solubility and other fate and transport characteristics are so
dramatically different (Table 2). Within each of these groups, the major
differences in fate and transport properties were related to the EC numbers of
the compounds. As shown in Table 2 and Figure 3, the properties of the
compounds (i.e., solubility, vapor pressure) and their modeled environmental
behavior [i.e., coefficient for partitioning to organic carbon (Koc) from soil to
water (LF sw)] change by an order-of-magnitude or more between the different
EC number fractions.
63
Chapter 4
Table 2. Fate and transport characteristics of TPH fractions (based on EC

number) [3].
Aliphatic Fractions
C5-C6
>C6-C8
>C8-C10
>C10-C12
>C12-C16
>C16-C35
Aromatic Fractions
C6-C7 (Benzene)
>C7-C8 (Toluene)
>C8-C10
>C10-C12
>C12-C16
>C16-C21
>C21-C35
Solubility
(mg/L)
Vapor
Pressure
(Atm)
Log Koc
H
(cm 3/cm 3)
LFsw
(mg/L)/
(mg/kg)
3.6E+01
5.4E+00
4.3E-01
3.4E-02
7.6E-04
1.5E-06
3.5E-01
6.3E-02
6.3E-03
6.3E-04
4.8E-05
1.1E-06
3.0E+00
3.6E+00
4.8E+00
5.9E+00
6.7E+00
8.6E+00
3.40E+01
5.10E+01
8.20E+01
1.30E+02
5.40E+02
1.10E+02
6.25E-03
1.73E-03
2.51E-04
3.26E-05
1.64E-06
8.26E-09
1.8E+03
5.2E+02
6.5E+01
2.5E+01
5.8E+00
5.1E-01
2.9E-02
1.3E-01
3.8E-02
6.3E-03
6.3E-04
4.8E-05
7.6E-06
1.6E-08
1.9E+00
2.4E+00
2.9E+00
3.2E+00
3.8E+00
4.2E+00
5.1E+00
2.25E-01
2.70-01
4.90-01
1.40-01
5.40E-02
1.20E-02
8.20E-05
9.00E-02
3.37E-02
5.16E-03
3.28E-03
1.64E-03
5.21E-04
6.56E-05
Koc = Carbon- water sorption coefficient

LFsw = Leaching factor from soil to water
H
= Henrys Law constant
Figure 3. Solubility vs. equivalent carbon number for aromatics and aliphatics
[3].
64
Chapter 4
Toxicity criteria were developed for each fraction as shown in Table 3. The
methodology used by the USEPA and the TPHCWG for developing the
reference doses (RfDs) and reference concentrations (RfCs) is described in
detail in Chapter 7.
Table 3. Toxicity criteria for TPH fractions [4].
Equivalent Carbon
Number Range
Aliphatic
<C6
>C6-C8
Aromatic
C6-C7 (Benzene)
>C7-C8 (Toluene)
Aliphatic
>C8-C1 0
>C10-C12
>C12-C16
Aromatic
>C8-C1 0
>C10-C12
>C12-C16
Aliphatic
>C16-C35
Aromatic
>C16-C35
Oral RfD
(mg/kg/day)
5.0
Inhalation RfC
(mg/m3)
18.4
0.2
0.4
0.1
1.0
0.04
0.2
2.0
NA
0.03
NA
NA = Not available
Adapting the TPHCWG Analytical Methodology to Crude Oils
The original TPHCWG approach does not include hydrocarbons greater than
carbon number 28 (C28), but has been modified to include up to C35 [3]. This is
appropriate for most refined products, such as gasoline and diesel, as well as for
condensates. However, some crude oils with very low API gravity values may
contain as much as 50 to 60% hydrocarbons >C35. Therefore, the TPHCWG
analytical methodology was further modified so that hydrocarbons up to C44
could be fractionated and detected by GC (as described in Chapter 5). Figure 4
presents the data for 15 different crude oil samples plus a diesel sample using
the extended GC analytical method. From this figure, it can be seen that
approximately 45 to 80% of the hydrocarbons in crude oil can be detected using
a gas chromatograph (i.e., EC numbers from C6 to C44).
65
Chapter 4
100
C6GC
to -C44
%
C6 to C44
<C6
>C45
% Detected by GC
90
80
70
60
50
40
30
20
10
0
5 6 7 8 9 10 11 12 13 14 15
Different Crude Oils
Diesel
Figure 4. Mass balance obtained for crude oils using a modified TPHCWG
analytical method and distillation data.
The remaining hydrocarbons fall into two groups: those <C6 or those >C44.
These ranges are typically quantified using distillation to determine the entire
composition of a crude oil. The hydrocarbon fraction with carbon numbers
greater than C44 is sometimes called the vacuum residuum, because it contains
the compounds remaining after the vacuum distillation of crude oil. The less
than C6 fraction is lighter than gasoline and is predominately the component of
natural gas. The amounts of hydrocarbons <C6 and >C44 (obtained from
distillation) for each crude oil are also shown in Figure 4. The addition of all
three molecular weight ranges accounts for more than 85% of the compounds
within crude oils. At the same time, greater than 95% of diesel oil can be
detected by the TPHCWG analytical method alone, further reinforcing that the
TPHCWG method adequately quantifies refined products such as diesel.
There are no standard USEPA analytical methods available for quantifying
either the <C6 or the >C44 fraction (vacuum residuum). Because the <C6 fraction
is generally lost to volatilization after a crude oil release, it is likely to be
unnecessary to quantify this fraction for assessing potential health risks for soil
exposures. On the other hand, the fate and transport characteristics of the >C44
fraction or vacuum residuum (see Table 4) indicate that it may remain in soils
even after extensive weathering of an oil has occurred.
The vacuum residuum fraction of a crude oil is comprised of very large
molecules (those boiling above 600C) that are not well characterized as to their
compositional make up, but it is known to contain a mixture of aliphatics,
aromatics, metals, and asphaltenes. Because of the complex nature, limited
66
Chapter 4
mobility, and the small amount of published toxicity data on this fraction, a
decision was made to evaluate it as a single fraction, rather than trying to
separate it into its aliphatic and aromatic components. The amount of vacuum
residuum in 800 crude oils from the United States is shown in Figure 5 [14].
80
70
60
50
40
30
20
10
0
0
10
20
30
40
50
60
70
API Gravity of Crude Oil

Yield (Vol %) of Vac Residuum
Figure 5. The yield of vacuum residuum in 800 crude oils produced in the
United States [14].
To determine the mass of vacuum residuum in a crude oil, three approaches can
be used: (1) using the known amount of vacuum residuum present in a crude oil
as determined by distillation; (2) estimating the amount of vacuum residuum
from the API gravity; and (3) assuming that all material not accounted for by
GC is vacuum residuum. The first method is probably the most reliable, but
distillation data are not always available and may be costly to obtain. The
second method provides a rough approximation since it uses the slope of the line
that is fit through the data shown in Figure 5, so that:
% Yield of Vacuum Residuum = -1.253(API Gravity) + 69.32
This equation has a R2 value of 0.66. The third method is probably the least
reliable, but may be an acceptable approach if neither distillation nor API
gravity data are available.
67
Chapter 4
Summary of Recommended Changes

to the TPHCWG for Crude Oil Analyses
Figure 6 presents a modification of the TPHCWG approach for the aliphatic and
aromatic carbon number fractions that can be used to conduct a risk-based
assessment of the TPH that is associated with crude oils. The major changes
made to the original carbon number fractions of the TPHCWG shown in
Figure 2 are as follows:
(1) The >C21 to C35 aromatic carbon number fraction was replaced by a
>C21 to C44 carbon number fraction.
(2) The >C16 to C35 aliphatic carbon number fraction was replaced by a
>C16 to C44 carbon number fraction.
(3) A >C44 carbon number fraction was added that included both
aliphatic and aromatic hydrocarbons.
Aromatics
Aliphatics
H
|
C
HC
CH
HC
CH
H H H H H H H H
|
|
|
|
|
|
|
|
H CCCCCCCC H
|
|
|
|
|
|
|
|
H H H H H H H H
C
|
H C H
|
H
>C6-C7 (benzene)
>C7-C8 (toluene)
>C8-C10
>C10-C12
>C12-C16
>C16-C21
>C21-C44
>C5-C6
>C6-C8
>C8-C10
>C10-C12
>C12-C16
>C16-C44
>C44
Figure 6. Aliphatic and aromatic carbon number fractions for the assessment of
risk associated with crude oil TPH (highlighted fractions are different than
fractions of TPHCWG).
68
Chapter 4
The fate and transport chara cteristics for the C35-C44 and the >C44 fractions are
presented in Table 4. The toxicity characteristics were evaluated for these
fractions as described in detail in Chapter 7. Because of the fate and transport
characteristics of the C44 fraction, the only exposure pathway of concern will be
direct contact with surface soil.
Table 4. Fate and transport characteristics of additional TPH fractions (based
on equivalent carbon number) [11].
Aliphatic Fractions
>C16-C44
Aromatic Fractions
>C21-C44
Vacuum Residuum
>C44
Vapor
Solubility Pressure koc
(mg/L)
(Atm)
(cm 3/g)
H
(cm 3/cm 3)
LFsw
(mg/L)/
(mg/kg)
1.3E-06
7.6E-06
1.00E+09
6.40E+03
8.26E-09
6.6E-03
4.4E-09
1.26E+05
6.80E-04
6.56E-05
1.0E-04
NA
5.01E+05
4.10E-08
1.65E-05
Comparison of Crude Oil Composition

With Some Petroleum Products
Figure 7 provides a comparison of the distribution of the carbon number
fractions for a single analysis of four different mixtures of hydrocarbons (i.e.,
samples of gasoline, diesel, Vaseline , and a 34API gravity crude oil). Not
surprisingly, the gasoline is dominated by the lower carbon number aliphatic and
aromatic hydrocarbon fractions (>C6 to C10 aromatics and >C6 to C8 aliphatics).
On the other extreme is the Vaseline which consis ts almost exclusively of the
aliphatic fraction, >C16 (baby oil is similar in composition). Lastly, as expected,
the hydrocarbon fractions in a crude oil cover the full range of carbon numbers
for both the aliphatic and aromatic fractions.
69
Chapter 4
800000
Concentration (mg/kg
700000
600000
500000
Diesel
Gasoline
400000
Vaseline
Crude oil
300000
200000
100000
>C
44
be
nz
en
e
>8
tol
-1
ue
0
n
e
Ar
om
>1
0ati
12
cs
Ar
om
>1
2ati
16
cs
Ar
om
>1
ati
6cs
21
Ar
om
ati
cs
>2
1
Ar
om
ati
cs
>6
-8
Al
iph
>8
ati
-1
cs
0
Al
iph
>1
a
0tic
12
s
Al
iph
>1
2-1
at
i
cs
6
Ali
ph
ati
cs
>1
6A
lip
ha
tic
s
TPH Fractions
Figure 7. Comparison of the distribution of carbon number fractions in crude

oil and selected products.
Analytical Results for Crude Oils

As part of a joint industry Petroleum Environmental Research Forum project
(PERF 97-08), approximately 70 crude oils were analyzed using the modified
TPHCWG fractionation method. These oils were selected to cover a wide range
of API gravity values and geographical locations, and were contributed by
Chevron (33 oils), Exxon (15 oils), Shell (17 oils), and Unocal (5 oils). Figure 8
illustrates the sampling locations for 70 crude oils, with 30 of these oils being
from North America. The API gravity for these crude oils range from 8.8 to
46.4.
70
Chapter 4
10
16
3 1
1
8
2
53
3
33
4
1
Figure 8. Number of crude oil samples (70 total) analyzed by the geographic
region from which they originated [11].
One of the goals of the PERF study was to analyze enough crude oils so that the
results would be representative of all the crude oil types produced around the
world. Overall, crude oils from around the world exhibit a wide range in
molecular composition. However, if crude oils are classified on the basis of
aliphatic, aromatic, and non-hydrocarbon abundance, then the range in
compositional variation is limited due to the similarity in organic matter types
forming crude oil and predictable compositional changes that occur during crude
oil alteration. To illustrate this point, Tissot and Welte (1978) plotted the
composition of 636 crude oils from around the world and showed that crude oil
compositions fall in a narrow band extending from thermally mature crude oils,
enriched in aliphatic hydrocarbons, to biodegraded crude oils, with low amounts
of aliphatic hydrocarbons [15]. Figure 9 shows the composition of 51 crude oils
and six oily soils from the PERF study overlain on the global distribution for
crude oils from the Tissot and Welte study, and indicates that the analyzed oils
adequately represent the worlds oils.
71
Chapter 4
Aromatics 100%
l
95% Frequency
Distribution Envelope
World Oils
50%
100%
Saturates
PERF 97-08 Study Samples

(51 crude oils and crude oil
extracts from 6 soils)
50%
50%
100%
Resin + Asphaltenes
Figure 9. Triangular plot comparison of PERF oil composition to that of 636

oils from around the world.
The crude oil composition data obtained from the modified TPHCWG
fractionation method are shown in Table 5 for 70 crude oils. The >C44 data were
obtained for a smaller set of the crude oils. The >C44 data were obtained by
either vacuum distillation or were calculated using the API gravity correlation to
vacuum residuum. Table 5 also contains the results of USEPA Method 8260 for
volatile organic chemicals. The benzene and toluene values obtained by Method
8260 are presented rather than the C6-C7 and C7-C8 aromatic data obtained by
the TPHCWG analytical method because they are considered more reliable. As
discussed in Chapter 11, benzene is usually evaluated separately in risk-based
decision-making because it is a carcinogen. The polyaromatic hydrocarbon
compositions of 60 crude oils are presented in Chapter 10.
72
Chapter 4
Table 5. TPH fraction composition of 70 crude oils .

Standard Number Total
Mean Minimum Maximum Deviation of Non- Sample
(mg/kg) (mg/kg)
(mg/kg)
(mg/kg)
Detects Points
TPH Fractions
>C6-C8 Aliphatics
>C8-C10 Aliphatics
>C10-C12 Aliphatics
>C12-C16 Aliphatics
>C16-C44 Aliphatics
>C8-C10 Aromatics
>C10-C12 Aromatics
>C12-C16 Aromatics
>C16-C21 Aromatics
>C21-C44 Aromatics
>C44
Volatiles EPA8260
n-Hexane
Benzene (>C6-C7
Aromatics)
Toluene (>C7-C8
Aromatics)
Ethylbenzene
Total Xylenes
36,000
48,000
36,000
83,000
200,000
11,000
9,600
31,000
39,000
85,000
230,000
24.5
760
4,100
14,000
18,000
68
410
10,000
7,200
1,600
25
220,000
140,000
73,000
180,000
410,000
48,000
31,000
94,000
72,000
220,000
570,000
37,000
36,000
18,000
37,000
74,000
10,000
5,600
16,000
15,000
54,000
160,000
3
0
0
0
0
0
0
0
0
0
1
70
70
70
70
70
70
70
70
70
70
41
4,900
1,300
73
0.16 *
16,000
5,900
4,300
1,600
0
3
15
71
4,500
30
25,000
5,700
71
1,100
6,500
1.9
7.3
4,600
27,900
1,023
6,600
3
0
71
71
This value represents one-half the detection limit for benzene.
Analytical Results for Gas Condensates

Fourteen condensate samples were also analyzed by the modified TPHCWG
analytical method as part of the PERF study. Ten samples were contributed by
Arthur D. Little, Inc., and four were contributed by the Gas Technology
Institute. The API gravity for these condensates range from 45 to 70.1.
The composition of the condensates is shown in Table 6. No hydrocarbons >C44
were detected. No aromatic hydrocarbons >C21 were found in 9 of the 14
condensates. Table 6 also contains the results of USEPA Method 8260 for the
volatile organic chemicals: benzene, toluene, ethylbenzene, and xylenes. The
polyaromatic hydrocarbon data for 10 condensates are presented in Chapter 10.
73
Chapter 4
Table 6. TPH fraction composition of 14 gas condensates .

Standard Number Total
Mean Minimum Maximum Deviation of Non- Sample
(mg/kg) (mg/kg)
(mg/kg)
(mg/kg)
Detects Points
TPH Fractions
>C6-C8 Aliphatics
>C8-C10 Aliphatics
>C10-C12 Aliphatics
>C12-C16 Aliphatics
>C16-C44 Aliphatics
>C8-C10 Aromatics
>C10-C12 Aromatics
>C12-C16 Aromatics
>C16-C21 Aromatics
>C21-C44 Aromatics
>C44
Volatiles EPA8260
Benzene (>C6-C7
Aromatics)
Toluene (>C7-C8
Aromatics)
Ethylbenzene
Total Xylenes
218,231
154,296
74,117
69,778
44,936
60,577
19,837
15,633
6,089
3,707
56,000
85,000
569
ND
ND
24270
ND
ND
ND
ND
ND
670,394
230,430
130,000
210,000
200,000
117,399
29,000
43,000
28,000
20,000
ND
48,000
46,550
37,000
65,000
63,000
29,000
7,500
13,500
9,800
7,100
0
0
0
1
2
0
1
1
4
9
14
14
14
14
14
14
14
14
14
14
14
14
9,500
2,500
24,000
7,900
10
31,000
14,000
53,000
15,000
10
5,000
33,000
1,400
8,200
6,700
61,000
2,000
18,000
0
0
10
10
Summary
Most TPH analytical methods are not adequate for assessing potential risk to
human health because they are non-specific and provide little information on the
types of compounds present in terms of class or carbon range. The TPHCWG
developed a risk evaluation approach for complex petroleum hydrocarbon
products, based on an analytical method that divides the complex mixture into
several distinct carbon number fractions coupled with standard risk assessment
equations for assessing human exposure. With slight modifications, the
TPHCWG analytical method can be used to assess the hydrocarbon content of
crude oils. These modifications include extending the TPHCWG analytical
method to C44 and determining the fraction >C44 using one of three methods.
However, some regulatory agencies may opt to omit hydrocarbons >C44 in risk
assessments due to their lack of mobility in the environment.
TPH fraction data derived using the modified TPHCWG methodology can be
used to estimate potential non-carcinogenic human health risks. Since crude oils
vary widely in composition, 70 different crude oils were analyzed that were
representative of all the types of crude oils produced around the world. Data for
the equivalent carbon fractions and BTEX were presented. TPH RBSLs derived
using these data are presented in Chapter 8. Chapters 10 and 11 present more
74
Chapter 4
details on risk analysis of the indicator carcinogenic hydrocarbons, benzene and

some polyaromatic hydrocarbons.
References
1.
Weisman, W., Analysis of Petroleum Hydrocarbons in Environmental

Media, TPHCWG Vol. 1, Amherst Scientific Publishers, Amherst,
Massachusetts (1998).
2.
Potter, T. L., and Simmons, K.E., Composition of Petroleum Mixtures,

TPHCWG Vol. 2, Amherst Scientific Publishers, Amherst, Massachusetts
(1998).
3.
Gustafson, S., Tell, S.G., and Orem, D., Selection of Representative Total
Petroleum Hydrocarbon (TPH) Fractions Based on Fate and Transport
Considerations, TPHCWG, Vol. 3, Amherst Scientific Publishers,
Amherst, Massachusetts (1997).
4.
Edwards, D.A., Andriot, M.D., Amoruso, M.A., Tummey, A.C., Bevan,

C.J., Tveit, A., and Hayes, L.A., Development of Fraction Specific
Reference Doses (RfDs) and Reference Concentrations (RfCs) for Total
Petroleum Hydrocarbons (TPH), TPHCWG, Vol. 4, Amherst Scientific
Publishers, Amherst, Massachusetts (1997).
5.
Vorhees, D., Gustafson, J., and Weisman, W., Human Health Risk-Based
Evaluation of Petroleum Contaminated Sites: Implementation of the
Working Group Approach, TPHCWG, Vol. 5, Amherst Scientific
Publishers, A mherst, Massachusetts (1999).
6.
ASTM (American Society for Testing and Materials), Standard Guide for
Corrective Action Applied at Petroleum Release Sites, ASTM E 1739-95,
November 1995.
7.
Massachusetts Department of Environmental Protection (MADEP) Bureau

of Waste Site Cleanup and Office of Research and Standards, Background
Documentation for the Development of MCP Numerical Standards (1994).
8.
Louisiana Department of Environmental Quality (LDEQ ), Risk

Evaluation/Corrective Action Program, June 20, 2000.
9.
Texas Natural Reduction Conservation Commission (TNRCC), Texas Risk

Reduction Program (TRRP) Rule, adopted September 2, 1999, Texas
Register (24 TexReg 2208).
10. McMillen, S.J., Magaw, R.I., and Spence, L.R., Developing Risk-Based
Cleanup Standards for Soils in Nigeria, SPE Paper 66513, presented at the
SPE/EPA/DOE Environmental Conference, San Antonio, Texas (2001).
11. McMillen, S.J., Magaw, R.I., Kerr, J.M., Sweeney, R.E., Nakles, D.V., and
Geiger, S.C., A New Risk-Based Approach to Establish Cleanup Levels
75
Chapter 4
for Total Petroleum Hydrocarbons, presented at the International

Petroleum Environmental Conference, Houston, Texas, November 1999.
12. Hunt, J.M., Petroleum Geochemistry and Geology, San Francisco, W.H.
Freeman and Company (1979).
13. United States Environmental Protection Agency (USEPA), Test Methods
for Evaluating Solid Waste, Physical/Chemical Methods. Methods 3611
and 3630, Revision 1, SW -846 Manual, December 1996
14. Coleman, H.J., Shelton, E.M., Nichols, D.T., and Thompson, C.J.,
Analysis of 800 Crude Oils From United States Oilfields, Department of
Energy Report No. BETC/RI-78/14, November 1978.
15. Tissot, B.P., and Welte, D.H., Petroleum Formation and Occurrence,
Berlin: Springer-Verlag (1978).
76
Chapter 5
Performance of TPHCWG Analytical

Method for Crude Oils in Soils
Gregory S. Douglas,1 George Naughton,1 Ileana Rhodes,2 and Jill M. Kerr3
1
Arthur D. Little, Inc.

3
2
Introduction
In evaluating the application of the TPHCWG analytical method (see Appendix)
to crude oil and crude oil impacted sites, several important factors affecting
method performance were considered. The factors investigated included soil
type, oil characteristics (unlike refined products, crude oils vary greatly),
extraction procedures, fractionation procedures (separation of the extract into
aliphatic and aromatic hydrocarbons), and chromatographic conditions. Each of
these factors is described in the paragraphs below.
Soils
Soil types investigated were s and, loam, and clay. Table 1 summarizes the soil
samples and their characteristics. The petroleum hydrocarbons that were added
to the soils included aged hydrocarbons (residual from historic releases) and
fresh oil added at the laboratory.
Table 1. Soil and hydrocarbon type used to evaluate the TPHCWG method.
Sample
Soil Type
Previous Oil
Lab Spike
(% by Weight)
Oil Type
(API)
1
2
3
4
5
6
Sand
Sand
Loam
Loam
Clay
Clay
Remediated <1% TPH

None
None
None
Remediated <1% TPH
Remediated <1% TPH
2%
1.5%
4%
2%
3%
1%
38
11
15
34
25
22
Oils
Six different crude oils, in addition to the historic, aged hydrocarbons, were used
for the study. The American Petroleum Institute (API) gravity (o) of the oils
ranged from 11o to 38o. Oil was spiked into each soil sample to achieve a TPH
77
Chapter 5
concentration of between 1.5 and 4% oil by weight. Total recovery of oil was
calculated against an identical direct analysis of the oil. Heavier oils, with lower
API gravity values, tend to have greater proportions of >C28 material, while
some lighter oils, with higher API values, tend to have greater proportions of
paraffins (normal alkanes). These characteristics may bias the analysis of crude
oils using the current TPHCWG method.
Extraction
Four extraction procedures were investigated. If different extraction procedures
provided comparable results, then commercial laboratories would have greater
freedom in meeting a performance-based criteria method, and data users would
have greater opportunity to compare data from different sources.
The TPHCWG method is a performance based method that

recommends the extraction of 10 g of soil (dried with NaSO4) with
10 mL of pentane by vortexing or shaking for a minimum of 2 minutes
(sometimes an overnight shaker is used). An aliquot is withdrawn from
the final extract for analysis without concentrating the sample.
The TPHCWG method was modified to use methylene chloride instead

of pentane. This requires the additional step of exchanging the extract
to pentane before column fractionation; this also involves minimal
concentration of the extracted material.
Soxhlet extraction (USEPA Method 3541 [1]) requires the extraction of

a 5 to 10 g soil sample (dried with NaSO4) with 100 mL of
heated/boiling solvent (usually methylene chloride); this procedure
requires extract concentration to maintain detection limits.
Ultrasonic extraction (USEPA Method 3550B [1]) employs sequential

extractions of soil with room temperature solvent (e.g., methylene
chloride). This method uses the largest volume of solvent, thus
requiring the most concentration, usually performed by KudernaDanish (KD) concentration. This method has the advantage of
increased oil recovery because of the sequential extractions. Sequential
extractions could be applied to other methods for the same purpose;
however, transfer and volatile losses are possible.
Fractionation
Four different fractionation (separation of the extract into aliphatic and
aromatic) procedures were evaluated. Two procedures are provided in the
TPHCWG method based on modified USEPA Methods 3630 Silica Gel and
78
Chapter 5
3611 Alumina Separations [1]. The other two procedures are modifications of
USEPA Method 3630 [1], Silica Gel Separation.
The fractionation methods were evaluated in two ways. First, a laboratory
standard containing normal alkanes, benzene, toluene, ethylbenzene, and
xylenes (BTEX), and the 16 priority pollutant polyaromatic hydrocarbons
(PAHs) was prepared and fractionated by each procedure. Second, direct
analysis of non-fractionated oils and extracted soil residues were compared to
post-fractionation data to monitor mass balance (aliphatic/aromatic ratios) and
extraction recovery efficiency.
The TPHCWG provides two procedures for the separation of hydrocarbons into
aliphatic and aromatic class based extracts. These employ alumina or silica gel
chromatography. Both methods use small volumes for ease, expense, and to
alleviate the need to concentrate extracts. The TPHCWG procedures plus the
two additional procedures are summarized below.
Alumina chromatography uses 4 g activated alumina as the solid phase

and pentane to elute the aliphatic fraction and methylene chloride for
the aromatics.
Silica gel chromatography uses 2 g activated silica gel (75-250 mesh)

as the solid phase and pentane (to elute the aliphatic fraction) and 1:1
acetone:methylene chloride (to elute the aromatic fraction) as the
mobile phase, or eluents.
USEPA Method 3630 [1] was modified to use 11 g silica gel as the
solid phase and pentane (to elute the aliphatics) and 1:1
pentane:methylene chloride (to elute the aromatics). This method
resulted in greater volumes of solvent being used that would require
extract concentration to maintain detection limits.
USEPA Method 3630 [1] was further modified by using a high

performance liquid chromatographic system to optimize the
aliphatic/aromatic separatio n. This method used a packed silica gel
column with a dual-solvent gradient elution program using pentane and
methylene chloride. Sample collection windows for the aliphatic and
aromatic compounds were determined using aliphatic and aromatic
control standards.
Chromatographic Conditions
Chromatographic conditions for the gas chromatography with flame ionization
detection (GC/GFID) described in the TPHCWG method were designed for use
with refined petroleum products such as gasoline and diesel. Thes e conditions
79
Chapter 5
may not be optimal for analysis of crude oils and were examined in this study.
Chromatographic conditions affecting the analysis of crude oil that were
investigated include loading capacity, effective carbon elution range, and mass
discrimination.
Results
This study was designed to evaluate the impact of sample processing on the
analysis of crude oil for TPHCWG method results. The critical elements in the
performance of this method were identified as extraction, fractionation of the
extract into aliphatics and aromatics, and GC/FID separation (chromatography).
Extraction
Table 2 compares the total recoverable petroleum hydrocarbon results for the six
soil types and four extraction techniques. Pentane does appear to be an adequate
solvent for extracting crude oils from soil, achieving recoveries similar to those
for methylene chloride. However, these results further demonstrate that TPH is
method defined, with each solvent and extraction procedure giving somewhat
different results. Laboratories should be flexible in their performance of the
method and evaluate potential bias by analyzing a matrix spike and matrix spike
duplicate with each analytical batch.
Table 2. Total recovered petroleum hydrocarbons (mg/kg).
Method
Concentration of
Added Oil in Soil
Vortex Pentane
Vortex Methylene
Chloride
Soxhlet Methylene
Chloride
Sonication
Methylene Chloride
1-Sand
30,000
2-Sand
15,000
Sample
3-Loam
4-Loam
40,000
20,000
5-Clay
40,000
6-Clay
20,000
24,700
23,200
8,160
7,500
21,700
33,700
15,800
16,400
30,600
43,000
7,250
11,200
20,800
7,170
18,100
15,100
41,100
14,400
NA
6,730
19,400
14,900
45,800
20,800
NA Sample was lost in lab and could not be replaced.
Fractionation
All four fractionation procedures were evaluated using a laboratory standard
containing normal alkanes, BTEX, and the 16 priority pollutant PAHs. As with
all fractionation procedures, it is imperative that the effectiveness of the
separation column be tested prior to any sample analysis. Several factors
influence the efficiency of a fractionation column to resolve compound classes.
They include activation/deactivation of the silica or alumina, type of solvents,
type of oil, and column overloading. Potential deactivation of the pre-packaged
80
Chapter 5
columns with moisture is a potential problem and they are therefore not
recommended.
For comparison purposes, GC/FID chromatograms for the TPHCWG and
Modified USEPA Method 3630 fractionation procedures are provided in Figures
1a and 1b, respectively. Figure 1a shows some compound class carryover
observed in the TPHCWG fractionation procedure which demonstrates the need
for the laboratory to monitor the fractionation efficiency carefully. When this
situation is encountered, the column must be reactivated before any samples are
analyzed.
F1 (aliphatic)
Response_
Note bleed of light

aromatics (BTEX)
into F1 fraction
60000
55000
11199714.D\FID1A
50000
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Time
F2 (aromatic)
Response_
11199725.D\FID1A
40000
35000
Irregular
distribution of
PAHs indicates
problem
30000
25000
20000
15000
10000
5000
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Time
Figure 1a. The sample fractionation procedure mu st be carefully monitored

with aliphatic and aromatic standards. This is an example of a separation
performed using pre-packaged columns; less than 1 mg total material was
loaded onto the column. The column was either improperly activated or
overloaded.
81
Chapter 5
F1 (aliphatic)
Response_
No aromatics
observed in F1
fraction
55000
50000
11199736.D\FID1A
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Time
F2 (aromatic)
Response_
11199747.D\FID1A
50000
45000
40000
35000
30000
25000
20000
15000
10000
5000
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Time
Figure 1b. Fractionation should be monitored with a check standard. (This

separation performed following modified USEPA Method 3630, Silica Gel
Separation.)
Figure 1b shows proper frationation with minimal compound class carryover.
When proper activation is achieved, no problems are encountered with the
modified methods using columns prepared by the laboratory . The two
modifications of USEPA Method 3630, both open column and HPLC, provided
comparable results. These procedures employ available technology and are
easily performed by most environmental laboratories. The performance-based
nature of this method should allow laboratories to produce comparable results
through careful modification of USEPA Method 3630. Table 3 compares the
results of the four fractionation procedures on two soil samples.
Table 3a. High API gravity oil in sand.
Total
Aliphatic
Method
(mg/kg)
TPHCWG Alumina
13,000
TPHCWG Silica
12,000
USEPA 3630 Silica
11,000
USEPA 3630 Silica HPLC 11,000
Total
Aromatic
(mg/kg)
3,100
5,000
5,300
4,700
%
Aliphatic
81
70
67
70
82
Chapter 5
Table 3b. Low API gravity oil in sand.
Method
TPHCWG Alumina
TPHCWG Silica
USEPA 3630 Silica
USEPA 3630 Silica HPLC
Total
Aliphatic
(mg/kg)
5,400
4,900
4,200
4,100
Total
Aromatic
(mg/kg)
3,700
4,000
4,100
4,000
%
Aliphatic
59
55
51
50
As with any fractionation method for compound class separation, it is essential

for laboratories to evaluate and confirm that the fractionation method used is
adequate for crude oils by using a column calibration standard. The standard
used in this analysis (or a similar one) should be adopted by method users as a
requirement for the demonstration of method performance.
Visual analysis of the aliphatic and aromatic fractions of crude oils can help
identify problems with the fractionation procedures. The aliphatic fraction has a
typical alkane distribution characteristic of this compound class, as shown in
Figure 2a.
(a) Aliphatic Fraction
83
Chapter 5
(b) Aromatic Fraction
Figure 2. Aliphatic (a) and Aromatic (b) fractions of high API gravity crude oil.
However, in biodegraded oils (Figure 3a, low API gravity) this aliphatic
hydrocarbon pattern may not be as evident due to the biodegradation of the nalkanes. In the case of the aromatic fraction, the high API gravity oil
demonstrates a characteristic pattern of one, two, and three ring aromatics
(Figure 2b); however, in the low API gravity oil the characteristic pattern is
degraded and replaced by an unresolved aromatic complex mixture (Figure 3b).
Identification of carryover in the low API gravity oil is more difficult than the
more common lighter crude oils, therefore careful analysis of quality control
samples is critical for the successful application of this procedure. For waxy
crude oils, the carryover of high molecular weight n-alkanes is a frequent
problem that can only be identified by evaluating the chromatograms.
84
Chapter 5
(a) Aliphatic
(b) Aromatic Fraction
Figure 3. Aliphatic (a) and Aromatic (b) fractions of low API gravity crude oil.
Chromatographic Conditions
The original version of the TPHCWG analytical methodology did not include
hydrocarbons greater than carbon number 28 (C28). This is appropriate for most
refined petroleum products such as gasoline and diesel. However, the
concentration of hydrocarbons with carbon numbers greater than 28 (or even 35,
the upper limit for the current TPHCWG analytical method) can be as high as 50
to 60% in some crude oils with low API. Therefore, to conduct a better risk-
85
Chapter 5
based analysis of sites where crude oil is present, it may be necessary to be able
to detect hydrocarbons with carbon numbers greater than C35. This can be done
by modifying the gas chromatographic technique to quantify hydrocarbons up to
C44. (C4 4 is the upper limit for most gas chromatographs, and C44 is therefore
the technical limit of the instrument. Hydrocarbons >C44 may be present in
crude oils as described in Chapter 4.) The chromatographic conditions necessary
to do this are not routine in most environmental laboratories.
Because the TPHCWG analytical method is chromatography based, it is
imperative that the chromatographic system not introduce large amounts of bias.
In this analysis, the most common source of bias was found to be mass
discrimination [3]. Mass discrimination is created when the chromatographic
system, be it the chromatographic column or the detector, responds differently to
different materials based on their relative mass. In the TPHCWG method, the
use of a flame ionization detector (FID) ensures that discrimination does not
occur at the detector end of the system. In general, the response of an FID is
essentially the same for all hydrocarbons (on a weight basis). Saturated
hydrocarbons and compounds containing heteroatoms, such as benzothiophene,
have a slightly lower response than unsaturated, aromatic hydrocarbons because
of the lower carbon to hydrogen and carbon to heteroatom ratio of these
compounds.
Another source of mass discrimination can be created by injection port
conditions if those conditions are preferential to a particular class or classes or
compounds. In this case, the use of a narrow bore column [0.25 m innerdiameter (ID)] creates an injection port environment preferential to smaller,
lower boiling point compounds. Larger, heavier, higher boiling point
compounds may not enter the chromatographic column at the same rate and may
be underestimated. The use of a wider bore column (0.32 m ID) and proper
positioning of the column in the injection port can alleviate much of the mass
discrimination. Mass discrimination of less than 20% between n-C20 and n-C40
can be easily maintained. When analyzing crude oils these performance criteria
should replace that in the TPHCWG method comparing n-C20 to n-C28, as this
range is more applicable to the analysis of crude oils. Special attention and
maintenance must also be applied to the injection port.
The use of a wider bore column (e.g., 0.32 m ID) can alleviate many mass
discrimination problems, but can present new problems as well. The use of a
wider bore column will reduce the separation of early eluting compounds,
require a reduced temperature ramp rate, and result in longer analysis times. The
recommended GC conditions for crude oil analysis are as follows:
86
Chapter 5
Gas Chromatographic Conditions

Carrier Gas:
Carrier Gas Flow:
Air Flow:
Hydrogen Flow:
Make-up Gas Flow:
FID Temperature:
Injection Port Temperature:
GC Operation:
Straight Liner:
Column Head Pressure:
Linear Velocity:
Oven Program:
Column:
Helium
5 to 7 mL/Min.
360 mL/Min.
35 mL/Min.
30 mL/Min.
325C
275C
Splitless Mode
4 mm ID Packed With Glass Wool
12.0 psi @ 50C
Approximately 50 cm/sec
Initial Temperature 10C
Hold for 5 Min.
Ramp @ 4C/Min. to 320C
Hold for 10 Min.
DB-5 30 m x -0.32 m ID, 0.25 m Film
Thickness
Compound Specific Analyses Are Required

The TPHCWG has recommended using USEPA Method 8260 [1], a method
specific to volatile organics, for accurately measuring benzene and toluene
content of oils. The need to use USEPA 8260 to quantify volatiles was
confirmed in this study. In addition, action levels for benzene and toluene are
lower than can be achieved with the TPHCWG analytical method. The same
applies to the measurement of individual target PAHs which should be
determined using USEPA Method 8270 [1].
Summary
Analytical methods for determining the petroleum hydrocarbon content of soils
should be performance based, with common data quality objectives for elements
such as mass discrimination, mass balance, extraction efficiency, fractionation
efficiency, and calibration. When sites have been impacted by crude oils, oil
fractionation must be carefully determined and monitored, or risk-based
screening levels may be biased.
Benzene and toluene concentrations should be determined by a method specific
to volatile organics, such as USEPA Method 8260 [1]. These results can then be
combined with the TPH determinations for risk analysis.
87
Chapter 5
References
1.
United States Environmental Protection Agency (USEPA), Office of Solid

Waste and Emergency Response, Test Methods for Evaluating Solid
Waste (SW -846), Washington, D.C. (1986).
2.
Personal Communication from Ileana Rhodes, November 2000.
3.
Douglas , G.S., McCarthy, K.J., Dahlen, D.T., Seavey, J.A., Steinhauer,

W.G., Prince, R.C., and Elmendorf, D.L., The Use of Hydrocarbon
Analysis for Environmental Assessment and Remediation, Journal of Soil
Contamination, 1, 197-216 (1992).
88
Chapter 6
Predicting the Fate and Transport of

Hydrocarbons in Soil and Groundwater
Lynn R. Spence,1 Kirk T. OReilly,2 Renae I. Magaw,2 and William G. Rixey3
1
Spence Environmental Engineering

3
Department of Civil and Environmental Engineering, University of Houston
2
Introduction
Effective risk-based decision-making for chemicals that have been released to
the environment requires a basic understanding of the fate and transport of the
chemicals once released and how they interact with the different environmental
media (such as soil, water, air, and biota) that they contact. This information is
necessary in order to determine how people, or other receptors, might be
exposed to the chemicals and what the associated risks might be.
Exploration and production (E&P) sites are usually concerned with spills of
crude oils, condensates, or wastes that are associated with E&P activities.
Accordingly, spills of individual chemicals are rarely of concern and instead,
spills of hydrocarbon mixtures predominate. This is important to the evaluation
of the fate and transport of chemicals at E&P sites because the presence of the
hydrocarbon mixture will affect the solubility of any individual chemical and its
movement in the environment.
When a chemical is introduced to soil, it can volatilize, sorb to soil particles,
dissolve into the groundwater (or soil pore water), or remain in residual form.
This partitioning of the chemical into other media is described by a set of simple
mathematical equations. Movement of a chemical through different media
depends on the unique physical and chemical properties of the individual
chemical, the amount of the chemical in relation to the amount of total
petroleum hydrocarbons (TPH) in the spill, and the properties of the media. In
risk-based decision-making, this mo vement through the environment is typically
modeled with a variety of fate and transport models. Some basic equations that
can be used to estimate chemical transport will be presented in this chapter.
Many of the fate and transport models typically used for risk evaluation are
based on these equations.
The goal of this chapter is to provide an overview of the important fate and
transport processes that affect petroleum spills in soil, and the estimation of risk
from spills. The concept of partitioning when the chemical is part of a mixture
89
Chapter 6
(Raoults Law) is described, as are the potential implications of including

Raoults Law in fate and transport modeling of petroleum hydrocarbons.
Chemical Fate and Transport in

Risk-Based Decision-Making
As described in Chapter 1, risk-based decision-making (RBDM) for
environmental site management generally requires that a quantitative risk
evaluation be performed. The risk evaluation process includes four technical
elements, all of which are necessary for a complete evaluation of risk. The four
elements are:
Exposure Assessment
Toxicity (Dose-Response) Assessment
Chemical fate and transport processes are integral to completing the exposure
assessment. Figure 1 illustrates a simple conceptual site model (CSM) in which
potential exposure pathways are shown for an oil production site. Workers may
be exposed to surface soil, or to vapors from subsurface soil or groundwater,
while off-site residents may be exposed to dissolved chemicals in their drinking
water. Identifying exposure pathways is a critical component of risk-based
decision-making and the CSM is a very useful tool for visualizing the
exposure pathways that may exist at a particular site.
People
Exposure
Exposure
Impacted
shallow soil
Drinking
water
well
Groundwater
Hazard
Figure 1. Example of a conceptual site model at an oil production site.
90
Chapter 6
In order to quantitatively estimate the potential exposures that people might

have, it is necessary to first know how much of the chemicals of concern may be
present in the environment. This can be done either by actually measuring
chemical levels in the different environmental media (soil, water, and/or air) or
by estimating levels with mathematical models. Measurement of chemical
concentrations is most often used for exposure pathways in which the receptor is
directly exposed to the environmental medium of concern. Mathematical
models are typically used to estimate the transfer of contaminants from the
source of the chemical to the receptor for exposure pathways in which the
receptor is indirectly exposed.
Many risk-based environmental management programs focus on potential
human exposures to three environmental media: surface soil, subsurface soil,
and groundwater [1,2,3]. Each of these media has more than one associated
potential exposure pathway. This section describes the exposure pathways for
each medium and indicates whether they are direct or indirect.
Surface Soil
Surface soil is usually defined as the soil at the ground surface that could
directly contact a receptor. Most regulatory programs have defined the top
1-meter of soil to be surface soil, however some programs define this zone as
the top 3 meters for residential scenarios. The potential exposure pathways that
are associated with surface soil are:
Ingestion of Soil
Dermal Contact
Inhalation of Volatile Emissions in Outdoor Air
Inhalation of Dust Emissions in Outdoor Air
The first two exposure pathways are direct pathways and no modeling is
required to estimate the concentrations of chemicals of potential concern that
receptors may contact. The latter two pathways are indirect pathways in which
the chemicals of potential concern originate in the impacted soil, but then are
transferred to air as they volatilize. The concentrations of the chemicals of
concern transferred from soil to breathing zone air must be estimated using
mathematical fate and transport models.
Subsurface Soil
Subsurface soil is usually defined as the region directly beneath the surficial soil
extending to just above the water table. A receptor is not ilkely to directly
contact subsurface soil. However, soil in this unsaturated zone may act as a
source of contamination to the ground surface via volatilization or to
groundwater via leaching.
91
Chapter 6
Three different exposure pathways may be of concern:
Inges tion of Groundwater (From Leaching)

Inhalation of Volatile Emissions in Indoor Air
All three pathways are indirect and, therefore, require the use of fate and
transport models to estimate the concentrations of chemicals of potential
concern to which receptors may be exposed.
Groundwater
There are usually three groundwater exposure pathways:
Ingestion of Groundwater
Inhalation of Volatile Emissions in Indoor Air
Ingestion of groundwater is a direct exposure pathway. The two volatilization

pathways are indirect and therefore models are used to estimate volatilization
from groundwater through the vadose zone.
In Tier 1 of most state risk-based corrective action (RBCA) programs, the
groundwater risk-based screening levels (RBSLs), whether they are United
States Environmental Protection Agency (USEPA) drinking water maximum
contaminant levels (MCLs) or risk-derived numbers, are assumed to apply at the
source of the contamination and are therefore required to be met everywhere in
the aquifer. Some states, however allow the use of a buffer zone, which is
some distance downgradient from the source, at which the groundwater targets
must be met [4]. In the latter case, a transport model would be used to estimate
the transport from the groundwater source area to the downgradient groundwater
receptor exposure point.
Partitioning in the Source Area

The first step in modeling the fate and transport of a chemical ni soil and
groundwater is to describe the distribution of the chemical in the contaminant
source region. The source region is defined in this chapter as the soil that
contains the bulk of the contamination. For example, although the initial source
of contamination may have been failed storage tanks, for risk assessment
purposes the source is considered to be the contaminated soil surrounding the
tanks that may serve as an ongoing source for chemical transport to a receptor.
The chemical distribution among the possible phases and locations in the soil
pore spaces in the contaminant source region must be accurately described in
92
Chapter 6
order to model the fate of the chemical in the environment. This is important
because the chemical in the dissolved phase will move with the flow of
groundwater (if in an aquifer) or move downwards with infiltration (if in the
unsaturated zone). Similarly, only the chemicals in the vapor phase are assumed
to be able to readily diffuse and flow with the soil vapor.
Partitioning describes the relative concentration of a chemical in two or more
phases in contact with each other at equilibrium. The total mass of a chemical in
soil is the sum of the mass in the vapor phase, dissolved phase, sorbed to soil,
and any chemical in residual form (in the organic phase):
M T = M s + M w + M v + M org
(1)
where:
MT
Ms
Mw
Mv
Morg
=
=
=
=
=
Total mass of the chemical in the soil (kg)

Mass of the chemical sorbed to soil (including organic carbon) (kg)
Mass of the chemical dissolved in groundwater (kg)
Mass of the chemical in the vapor phase (kg)
Mass of the chemical in organic phase [as pure phase droplets or nonaqueous phase liquid (NAPL)] (kg)
Equation (1) would apply to the unsaturated zone where soil gas might be
present. If the mass of the chemical is in groundwater only, the mass would be
partitioned between the sorbed (sometimes called the solid phase), dissolved,
and organic phases:
M T = M s + M w + M org
(2)
where the parameters are as defined above. In Equations (1) and (2), Morg can
represent the mass of the pure form of the chemical, which occurs when a single
chemical is spilled, or it can represent the mass of the chemical found in a
separate-phase mixture, such as benzene in gasoline. The term, separate-phase
mixture is often called residual by professionals in the fields of risk
assessment and fate and transport. Used in this way, the term does not imply
mobility, but instead merely indicates the presence of a free product phase in the
soil. NAPL is also considered to b e residual product.
Equilibrium Partitioning Equation
Equations (1) and (2) define the components of the total chemical mass in soil,
however these equations dont show the relationships between the various
phases. Every chemical has a specific affinity for the various phases that can be
predicted with measured chemical properties and site-specific parameters. A
93
Chapter 6
chemical with a higher affinity for water will have a higher solubility than a
chemical with a lesser affinity for the water phase. Similarly, volatile
compounds have a high affinity for the vapor phase. In order to predict the
movement of the chemical in soil, the concentration in each phase must be
known. In the unsaturated zone, the concentration of a chemical in the soil can
be described as follows:
CT = ?bC s + ? wC w + ?a C v + ? orgCorg
(3)
where:
CT
Cs
Cw
Cv
Corg
=
=
=
=
=
a
w
org
Total concentration of chemical in soil (mg chemical/g dry soil)

Concentration of chemical sorbed to soil (mg chemical/g dry soil)
Concentration of chemical in water (mg chemical/cm3 water)
Concentration of chemical in vapor (mg chemical/cm3 vapor)
Concentration of chemical in the organic phase (mg chemical/cm3
residual)
= Air-filled porosity (cm3 air/cm3 dry soil)
= Water-filled porosity (cm3 water/cm3 dry soil)
= Porosity filled with residual phase liquid, either pure chemical (if a
single chemical was spilled) or the NAPL mixture (cm3 residual/cm3 dry
soil)
= Soil bulk density (g dry soil/cm3 dry soil)
The total concentration in soil, CT, is the concentration measured in the

laboratory through the analysis of a soil sample. For petroleum hydrocarbons,
the concentration in the organic phase, Corg, is the concentration of the
individual chemical in the oil.
For spills containing no residual, Equation (3) is written [5]:
CT = ?bC s + ? wC w + ? a Cv
(4)
Equation (4), often called the equilibrium partitioning equation, states that the
total concentration in soil can be partitioned among the sorbed, dissolved, and
vapor phases only; there is no residual present in the soil.
Many fate and transport models assume that the total concentration in soil can
be partitioned into three phases using Equation (4) [5]. In other words, the
models do not account for the separate oil phase. If the scenario being modeled
contains a NAPL phase, then Equation (4) will overestimate the concentrations
in each of the other three phases. Overestimation of the dissolved phase and soil
vapor phase will result in overestimation of the rate of leaching and
94
Chapter 6
volatilization. The following paragraphs describe how to determine whether the

use of Equation (4) is appropriate for fate and transport modeling of a specific
spill using the solubility of the chemical of interest.
Henrys Law Coefficient
Chemical partition coefficients are used to define the relationships between the
various concentrations in each phase. Henrys Law coefficient, KH, relates the
dissolved phase concentration to the vapor phase concentration in the following
manner:
KH =
Cv
Cw
(5)
where:
KH = Henrys Law coefficient for chemical [(mg/L vapor)/(mg/L water)]
Cw = Concentration of chemical in water (mg chemical/cm3 water)
Cv = Concentration of chemical in vapor (mg chemical/cm3 vapor)
KH has been measured for many chemicals and can be found in reference texts
[6,7]. It can also be estimated from the chemicals vapor pressure [8]. Higher
Henrys Law coefficients indicate chemicals with higher volatility. Values
range from 0.23 to 0.33 for benzene, toluene, ethylbenzene and xylenes (BTEX)
[7]. USEPA Region IX designates chemicals as being volatile if they have a
molecular weight less than 200 g/mole (mol) and a Henrys Law coefficient of
greater than 4.2E-4 (0.00042 units in dimensionless form) [9]. Table 1 shows
Henrys Law coefficients and other important fate and transport parameters for
BTEX.
Table 1. Chemical properties of BTEX [6,7].
Chemical
Parameters
BTEX:
Benzene
Ethylbenzene
Toluene
Xylenes
Molecular Vapor
Weight
Pressure
(g/mol)
(mmHg)
Henrys Law
Solubility Coefficient Koc
(mg/L)
(mL/g)
78.11
106.17
92.14
106.17
1770
169
530
198
95
9.6
28.2
8.06
0.227
0.328
0.276
0.293
6.6E+01
2.0E+02
1.4E+02
2.4E+02
The following equation is used to estimate the vapor phase concentration of a

chemical in soil from the dissolved phase concentration [8]:
95
Chapter 6
g
mg 1000 L 1000 g
C vapor 3 = K H C dissolved
m
L m 3 mg
(6)
Note, in Equation (6), Cv is in units of g/m3 and Cdissolved is in units of mg/L.

Both of these units are commonly used in risk assessment. Often soil vapor
concentrations are reported in units of parts per million-volume (ppmv); these
units must be converted to concentration in terms of mass per volume before
being used in risk assessment equations. For a gas at 25oC and 1 atmosphere
(atm) of pressure the conversion is:
g C ppmv MW
C vapor 3 =
24.45
m
106 g 10 3 L
3
g m
(7)
where:
Cppmv = Concentration of chemical in vapor reported as ppmv
MW = Molecular weight of chemical (g/mol)
24.45 = Molar volume at 25C and 1 atm (liter/mol)
RT
molar volume =
to calculate at other temperatures
P
R = Universal gas constant = 8.21E-2 [(atm liter)/(mol Kelvin (K))]
T = Temperature of the gas (K) (K = C + 273)
P = Pressure of the gas (atm)
Soil-Water Sorption Coefficient
The soil-water sorption coefficient, Kd, describes the relationship between the
concentration of the chemical sorbed to the soil to the concentration of the
chemical dissolved in water. The basic Freundlich equation is [6]:
Kd =
Cs
C wn
(8)
where:
Kd = Chemical-specific soil-water sorption coefficient [(mg chemical/g dry
soil)/(mg chemical/cm3 water)]
Cw = Concentration of chemical in water (mg chemical/cm3 water)
Cs = Concentration of chemical sorbed to soil (mg chemical/g dry soil)
n = Freundlich exponent (dimensionless)
96
Chapter 6
The sorption is usually assumed to be linear with respect to concentration in

water (i.e., n = 1). The soil-water sorption coefficient depends on both
chemical-specific and soil-specific properties. Most simple fate and transport
models, (e.g. Jury model [5,10], VADSAT model [11]), for organic chemicals
usually assume that Kd is equal to the product of the fraction of organic carbon,
Foc, and the organic carbon partition coefficient, Koc, of the chemical:
K d = Foc K oc
(9)
where:
Foc = Fraction organic carbon in soil (g organic carbon/g soil)
Koc = Chemical-specific organic carbon partition coefficient (mL/g)
The organic carbon partition coefficient can be found in many reference texts as
well as USEPA documents for specific chemicals [7,8]. The fraction of organic
carbon in soil is usually a site-specific value measured from a soil sample.
Equation (9) assumes that organic carbon is the sole predictor of chemical
sorption in soil. Clay content, the presence of charged particles and the fact that
organic carbon is somewhat difficult to measure, can cause the amount of
chemical sorbed to soil to be underpredicted, and hence the amount in the other
phases will be overpredicted. Lyman et al. [8] provides a discussion of the
increased sorption expected due to mineral content in the soil and provides some
equations for estimating the increased sorption. In some cases, it might be
important to have a better estimate of Kd than that provided by Equation (9). If
desired, Kd can be measured in the laboratory using soil gathered from the site.
Calculating Source Concentrations

Fate and transport models used for risk assessment usually begin with the
estimation of the chemical concentration in groundwater. If the model is used to
estimate volatile emissions, the vapor concentration is then estimated from the
dissolved phase concentration using Henrys Law coefficient.
For soil spills not containing a residual phase, the equilibrium partitioning
equation can be solved for the dissolved phase concentration by substituting
Equations (5) and (8) into Equation (4):
Cw =
? b CT
? b K oc Foc + ? w + ? a K H
(10)
Once Cw is calculated, the vapor phase concentration can be estimated:
97
Chapter 6
Cv = Cw K H
(11)
where the variables are as defined previously. Note, that the units of CT and Cw
are not the usual units reported in laboratory data. The conversions are as
follows:
mg
g
?b 3 CT
3
cm
mg
k g k g 1000 cm
Cw
=
L
L ?b K oc Foc + ? w + ? a K H 1000 g
(12)
In Equation (12), the soil concentration, CT, can be taken from the unsaturated
zone or below the water table. Equation (12) can be useful to predict the
concentration of a chemical in groundwater from a measured soil concentration,
however as written, it is not applicable for cases containing residual. If NAPL is
present (and it can be difficult to detect visually), Equation (12) will
overestimate the dissolved phase concentration. The presence of NAPL can be
determined by considering the chemicals solubility in water. If the calculated
dissolved phase concentration exceeds the chemicals solubility, then equation
(12) is not applicable. Since Equation (12) assumes that the chemical can be
partitioned into three phases (solid, dissolved, and vapor), then there must be a
fourth phase present and Equation (3) should be used. It is very important to
make sure that the fate and transport model includes the residual term (and
thereby limits the dissolved phase and vapor phase concentrations) if it is likely
that NAPL may be present in the soil.
Up until this point, the equations presented have been describing the behavior of
a chemical partitioning in soil as if it were the only chemical present. In
hydrocarbon mixtures, that is not the case and the chemicals solubility will be
limited even further due to the presence of the overall TPH mixture. Raoults
Law predicts the effective solubility for chemicals that are in mixtures and
accordingly, it should be used if the chemical being modeled is part of a
mixture. The next section defines Raoults Law.
Effect of Raoults Law on Source Concentrations
In fate and transport screening models, the partition equations described earlier
are usually applied in the source region to estimate the concentration of the
chemical that is partitioning from the source to the vapor and/or dissolved
phases. Fate and transport models typically consider one chemical at a time, and
the partitioning of the chemical between water, soil vapor, and soil organic
matter is calculated individually for each chemical. This is not an accurate
approach for mixtures of petroleum hydrocarbons. When a chemical is part of a
mixture, its partitioning behavior can change dramatically. This is especially
98
Chapter 6
true for TPH, where each chemical usually represents a very small percentage of
the overall mixture.
The accurate prediction of the concentration of an organic chemical contained in
a mixture of many organic chemicals is essential to the assessment of potential
risks. Accordingly, fate and transport equations used to examine petroleum
hydrocarbons should include this mixture effect. In order to model organic
chemicals as part of mixtures, the fate and transport equations require a minor
modification. The required modification is to use Raoults Law to account for
the chemicals effective solubility in water.
Effect of Raoults Law on Dissolved Concentrations
The maximum dissolved concentration of a chemical in an organic mixture is a
function of its pure compound solubility and the mole fraction of the chemical in
the organic phase. Published pure compound solubility values are available for
many compounds of concern and are listed in Table 1 for BTEX. When a
chemical is part of an organic mixture, its solubility in water will be limited as
follows:
Cw max = S eff = x * S
(13)
where:
Cw max = Maximum dissolved concentration of chemical in water (mg/L)
x
= Mole fraction of chemical in organic chemical mixture (mole
chemical/mole oil)
S
= Solubility of pure chemical in water (mg/L)
S eff = Effective solubility of chemical in water (mg/L)
Equation (13) is known as Raoults Law and it is commonly used to predict the
dissolved concentration of a chemical in water exposed to a hydrocarbon
mixture [8]. For example, benzene with a pure compound solubility of 1,770
mg/L and a mole fraction of 0.001 in crude oil would have a maximum
dissolved concentration of 1.77 mg/L. The maximum dissolved concentration,
when calculated in this manner, is also called the effective solubility. The mole
fraction is calculated from:
x=
CT MWTPH
MWi CTPH
(14)
where:
CTPH = Total concentration of total petroleum hydrocarbon in soil (mg/kg)
CT
= Total concentration of individual chemical in soil (mg/kg)
MWTPH = Average molecular weight of hydrocarbon mixture (g/mol)
99
Chapter 6
MWi
= Molecular weight of individual chemical (g/mol)
For chemicals with similar molecular weights, the mass fraction can be
substituted for the mole fraction. For example, if benzene were reported in a
soil sample at 100 mg/kg and the TPH concentration was reported to be 5,000
mg/kg, then the mass fraction would be = 100/5,000, or 0.02. Assuming the
molecular weight of benzene and the TPH mixture were similar (perhaps the
TPH is a light, refined product), the maximum dissolved phase concentration of
benzene expected would be equal to 1,770 mg/L x 0.02 = 35.4 mg/L (where
benzenes solubility = 1,770 mg/L).
Raoults Law is valid for compounds that are liquids at room temperature. For
solid compounds, such as the larger polycyclic aromatic hydrocarbons (PAHs),
the subcooled liquid solubility is used in place of the pure chemical solubility
(S) in Equation (13). These values are also available in the literature [12].
To confirm that Raoults Law is appropriate for predicting the effective
solubility of a chemical that is present in oil when the oil is in contact with
water, 15 crude oils were analyzed for 43 aromatic compounds ranging from
benzene through 5- and 6-ringed PAHs. The oil was then placed in contact with
water and compounds present in the oil were allowed to partition into the water.
The concentrations of the aromatic compounds in both the oil and water phases
were determined. Figure 2 presents the data for all analytes in which the
dissolved concentrations were above the analytical reporting limit of 5 ppb.
Most of the results were at or below the dissolved concentration predicted by the
application of Raoults Law. This is evidence that Raoults Law can be used to
estimate the effective solubility of a chemical.
100
Chapter 6
Measured Dissolved Concentration mg/L
1.0E+02
Measured
Raoult's
1.0E+01
1.0E+00
1.0E-01
1.0E-02
1.0E-03
1.0E-03
1.0E-02
1.0E-01
1.0E+00
1.0E+01
1.0E+02
Predicted Dissolved Concentration mg/L
Figure 2. Measured dissolved concentrations of analytes are generally less than

those predicted by Raoults Law.
Equation (13) predicts the effective solubility of a chemical in water from the
chemicals solubility, S, and mole fraction in the TPH, x. The concentration in
water can also be predicted using the chemicals oil-water partition coefficient if
known.
Cw = x S =
Co
Ko
(15)
where:
Cw = Dissolved phase concentration (mg/L)
Co = Concentration of the individual chemical in the TPH (mg chemical/kg
TPH)
Ko = Oil-water partition coefficient (L/kg)
x = Mole fraction of chemical in TPH (moles/mole)
Equation (15) can be used to calculate Ko for any organic phase if S and x are
known. Since x is a mole fraction, an estimate of the average molecular weight
of the organic phase is required.
Ko =
MWi
106 mg/kg
Si MWTPH
(16)
101
Chapter 6
where:
MW i = Molecular weight of the individual chemical (g/mole)
MW TPH= Average molecular weight of the TPH (g/mole)
S
= Solubility of the individual chemical (mg/L)
Based on the distribution of compounds in the following hydrocarbon products,
typical values for MWTPH in g/mole are:
Gasoline
Condensate
Diesel/Distillate
Crude Oil
Gas Oil
100
150
180
200-250
300
Effect of Raoults Law on Vapor Concentrations

Raoults Law is also applied when calculating vapor phase concentrations
because the vapor phase concentrations are calculated from the dissolved phase
concentration using Henrys Law coefficient [Equation (11)].
Soil vapor concentrations can also be calculated using the ideal gas law (PV =
nRT), where the mass of the chemical and its molecular weight have been
substituted for the number of moles, n:
Cv =
x MW VP
R T
(17)
where:
Cv
VP
x
R
T
= Concentration of chemical in the vapor phase (g/m3)

= Vapor pressure of chemical (mmHg)
= Mole fraction of chemical in liquid phase (underneath vapor) (-)
= Universal gas constant, 0.06236 [(mmHg m3)/(mol K)]
= Temperature in Kelvins (K) (293 K = 20C)
Equations (17) and (13) both indicate that the soil vapor concentrations are
limited by the presence of the TPH mixture because the mole fraction, x, will be
less than one. No matter how much NAPL is present in the soil, the vapor and
dissolved phase concentrations will not increase beyond their effective solubility
and effective saturated vapor concentration.
102
Chapter 6
Fate and Transport Processes

Fate and transport models are used to estimate concentrations in media other
than the source media for indirect exposure pathways. The estimated crossmedia concentrations can then be used to determine the exposure of a receptor to
the chemical of concern, via the exposure pathways described earlier in this
chapter.
The important transport processes addressed by these models include
volatilization, leaching, advection (being carried with the air or water), and
dispersion. For soil contaminants in the unsaturated zone, cross-media
concentrations are estimated for the following transport processes:
Leaching From Soil to Groundwater

Volatilization From Soil to Outdoor Air
Volatilization From Soil to Indoor Air
For contaminants in or on the groundwater, the transport processes that can be

modeled include:
Volatilization From Groundwater to Indoor Air

Volatilization From Groundwater to Outdoor Air
Transport in Groundwater to a Downgradient Location
The models used to estimate these transport processes and associated cleanup
levels do not address the movement of NAPL as a separate phase (e.g., the
vadose zone model SESOIL and the groundwater model AT123D [10]).
Groundwater dispersion models are used to estimate concentrations in
downgradient groundwater locations. These models are usually used in Tier 2
or Tier 3 risk analyses because the exposure point is located away from the
source. Some state regulatory programs, however, do include groundwater
transport to downgradient locations in Tier 1 [4].
The models commonly used for risk assessment tend to be screening models
because they are usually based on assumptions like uniform soil properties and
an unchanging groundwater flow field. Screening models are not applicable to
engineering design problems, such as the design of extraction wells, or for
complex hydrogeological flow regimes. Two examples of software that
integrate screening level fate and transport codes with risk assessment are BP
RISC and API DSS [11,10].
The next few sections of this chapter describe the key features usually found in
screening models.
103
Chapter 6
Air Models
If it is plausible that receptors could be exposed to soil emissions, then the
estimation of volatilization from the contaminant source will be important. If
the volatilization to the indoor air exposure pathway is being modeled, it often
drives cleanup levels because it can generate very low cleanup concentrations
due to deficiencies in the screening models. Screening models tend to
overestimate concentrations in air because they ignore many potentially
important processes impeding diffusion, they ignore biodegradation which can
be a significant loss mechanism for hydrocarbons, and the models are often
steady-state so they predict that the same exposure concentration will occur for
the entire exposure duration. For carcinogens like benzene, risk is a function of
total accumulated exposure and it is unlikely that a volatile emission exposure
would remain constant over a long exposure duration.
Diffusion
In screening models, the volatilization rate is usually estimated using the
following equation (from Ficks Law) (e.g., Farmers emission model [10,13],
groundwater emission model to indoor or outdoor air [11]):
E = Deff
(Cvs Cvf )
L
(18)
where:
E
Deff
Cvs
Cvf
L
= Volatilization rate toward the structure (g/cm2/sec)

= Effective diffusion coefficient in soil (cm2/sec)
= Vapor concentration at the source (g/cm3)
= Vapor concentration in the soil just outside the building foundation or at
ground surface (g/cm3)
= Volatilization distance (cm)
The effective diffusion coefficient in soil, Deff , is an extremely important input

parameter to the volatilization models. The effective diffusion coefficient is
calculated with the Millington-Quirk relationship and accounts for the tortuosity
in the soil [11,13]:
? 10/3
1 ?W 10/3
Deff = Dair a 2 + Dwater
?
K H ?T 2
T
(19)
where:
Dair
= Molecular diffusion coefficient for chemical in air (cm2/sec)
104
Chapter 6
Dwater = Molecular diffusion coefficient for chemical in water (cm2/sec)

TT
= Total porosity (cm3/cm3)
where the rest of the variables are as defined previously.
Diffusion coefficients in water are usually four orders of magnitude lower than
diffusion coefficients in air, so the second term in Equation (19) only becomes
important for cases with very low air content. Diffusion coefficients in air do
not tend to vary over a wide range so the sensitive input variable becomes the air
content of the soil. Because the models assume that the unsaturated zone is
homogeneous, Deff can greatly be overestimated. In reality, the presence of high
water content soil layers will greatly impede the diffusion of volatile chemicals.
It can be fairly simple to modify the screening models to include multiple soil
horizons with varying soil moisture contents [11,13]. Figure 3 shows the
sensitivity of calculated cleanup concentrations in subsurface soil with respect to
changes in air-filled porosity. The rest of the parameter values were chosen
from a typical Tier 1 analysis.
RBSL (mg/kg)
6
5
4
3
2
1
0
0
0.1
0.2
0.3
0.4
0.5
Air-Filled Porosity
Figure 3. Sensitivity of subsurface soil RBSL for indoor air to changes in airfilled porosity.
When estimating concentrations in outdoor air, the vapor concentration in the
soil at ground surface, Cvf, is assumed to equal zero (or is considered to be very
small compared to the vapor concentration at the source). For the indoor air
scenario, if pressure-driven flow is ignored, the vapor phase concentration in the
soil just outside the building foundation is also assumed to equal zero.
105
Chapter 6
Advection (or Pressure-Driven Flow)

Pressure-driven flow of the soil vapor may occur when a subsurface basement is
underpressurized with respect to the soil gas surrounding it. This situation may
arise due to wind loads on the building or heating systems operating in the
basement. In Tier 1 analyses, pressure differentials are not considered. Some
Tier 2 models include this term [11].
Mixing in Air
When using a screening model to estimate risk from volatile emissions, the
breathing zone is usually assumed to be a well-mixed box that is ventilated by
an air exchange rate. If the exposure pathway being considered is volatilization
to outdoor air, it is assumed that the receptor is directly over the source, and the
box is defined by the height of the breathing zone and the length of the source
in the predominant direction of wind flow. The width of the box
(perpendicular to the direction of wind flow) drops out of the equation. The
breathing zone height is usually assumed to equal 2 m. For indoor air, the box is
assumed to equal the volume of the building (for a house, it equals the volume
of the house), and the ventilation rate is the rate at which the air in the building
is exchanged with outdoor air.
Groundwater Models
The groundwater transport processes can be divided into those that occur in the
unsaturated zone and those that affect transport in an aquifer. In a Tier 1
analysis, the only unsaturated zone transport that is considered is the leaching
of the chemical of concern to the aquifer. In this case, the source in the
unsaturated zone is assumed to be located at the water table so there is no
opportunity for dispersion or degradation during transport to occur. Likewise, in
the saturated zone, the only transport process that is considered is the mixing
of the chemical in groundwater within a zone directly under the unsaturated
zone. Transport of the chemical from this mixing zone to a point downgradient
of the source is generally not considered in a Tier 1 analysis. As a result, the
accurate partitioning of the chemical among the phases and locations in the soil
pore spaces in the contaminant source region is essential for Tier 1 calculations.
Leaching From the Unsaturated Zone
Leaching from a source in the unsaturated zone is assumed to occur when water
in the form of precipitation infiltrates through the source, dissolves chemicals
and carries them to the water table. Screening models use the equilibrium
partitioning equations to predict the dissolved phase concentration in the source
region [2,10,11]. Usually the source is assumed to be uniform in concentration
and box-shaped.
106
Chapter 6
Under typical conditions, the amount of infiltration through the unsaturated zone
will not equal the sites average precipitation because of evaporation, uptake by
plants, and runoff. Many modelers assume a value in the range of 10 to 20% of
the average precipitation as an infiltration rate. For example, if the average
precipitation is 20 inches per year, a reasonable assumption for infiltration, or
the amount or water that will reach the water table, is about 2 to 4 inches per
year. The infiltration rate can be limited by the vertical hydraulic conductivity
of the vadose zone soil. Consideration should also be given as to whether the
source is underneath a paved area (thereby reducing infiltration) or in an area
with artificially high recharge (such as a truck wash area). The American
Petroleum Institute (API) has compiled a large database with recharge estimates
from around the United States. A brief technical report is available from API
summarizing the estimation of infiltration and the infiltration and recharge
database [14].
The mass loading from the soil to groundwater is calculated from:
Q = Cw A I
1000 cm3
(20)
where:
Q
Cw
I
A
= Contaminant mass flux at the water table (mg/yr)

= Dissolved phase concentration in leachate coming from source (mg/L)
= Infiltration rate (cm/yr)
= Cross-sectional area of source (cm2)
Besides being used in Tier 1 equations [1,2], Equation (20) is used by many
unsaturated zone models when they are linked with saturated zone models and
used to estimate concentrations downgradient [10,11].
The average downward seepage velocity of the water in the unsaturated zone is
calculated from [15]:
v=
I
?w
(21)
where:
v = Seepage velocity (cm/yr)

w = Water-filled porosity (cm3 water/cm3 soil)
107
Chapter 6
In reality, the hydraulic conductivity in the unsaturated zone changes as the

moisture content changes and water tends to move fast at first (a wetting front)
and then most likely becomes steady. Equation (21) represents an average
annual seepage velocity and is used by almost all simple analytical unsaturated
zone models because the screening models are being used to estimate long-term
loading to groundwater for purposes of calculating risk. This equation provides
a handy back-of-the-envelope reality check for questions such as: Could the
dissolved phase have reached the water table? and Is the travel time to the
water table so long that degradation will mineralize the chemical before it
reaches the groundwater?
Retardation (in Both Saturated and Unsaturated Zones)
Retardation describes a contaminants movement relative to the bulk movement
of groundwater flow. The retardation factor, R, is estimated using the following
equations for organic and inorganic chemicals:
?b FocK oc
?w
?K
R = 1+ b d
?w
R =1+
for Organic Chemicals
(22a)
for Inorganic Chemicals
(22b)
where:
Foc = Fraction organic carbon in dry soil (g organic carbon/g soil)
Koc = Organic carbon partition coefficient (mL/g or m3/kg)
Kd = Inorganic distribution coefficient (mL/g)
b = Soil bulk density of the saturated zone (g/cm3)
w = Water-filled porosity (cm3/cm3)
Retardation does not affect the results of screening models with constant source
terms (steady-state models) that ignore degradation. It only becomes important
when degradation is considered. This does not mean that the input parameters
used in the retardation equation are unimportant because these variables are also
used in the equilibrium partitioning equation, where they are very important.
Biodegradation
Biodegradation is usually assumed to be a first-order reaction in screening
models, which means that it is assumed to be dependent only on time. In most
models biodegradation is assumed to occur in the dissolved phase only, Cw (e.g.,
the groundwater models in the API DSS and BP RISC [10,11]). The sorbed
phases and NAPL are assumed not to degrade. This means that biodegradation
only begins as the contaminant migrates away from the source.
108
Chapter 6
The biodegradation rate is probably one of the most sensitive input parameters
for petroleum hydrocarbons, because the hydrocarbons that tend to be soluble,
and therefore mobile, are also highly degradable [16,17]. Biodegradation is
discussed in more detail in Chapter 11.
References
1.
Risk-Based Corrective Action Applied at Petroleum Release Sites, ASTM
E1739-5, November 1995.
2.
Risk-Based Corrective Action, ASTM E2081-00, April 2000.
3.

Program, August 1999.
4.
Colorado Department of Labor and Employment, Oil Inspection Section,

Petroleum Storage Tank Owner/Operator Guidance Document (1999).
5.
Jury, W.A., Russo, D., Streile, G., and El Abd, H, Evaluation of Organic
Chemicals Residing Below the Soil Surface, Water Resources Research, 1,
13-20 (1990).
6.
United States Environmental Protection Agency, Soil Screening Guidance:

Technical Background Document, Office of Solid Waste and Emergency
Response, EPA/540/R-95/128, May 1996.
7.
Howard, P.H., and Meylan, W.M., Handbook of Physical Properties of

Organic Chemicals , CRC Press, Boca Raton, Florida (1997).
8.
Lyman, W.J., Reidy, P.J., and Levy, B., Mobility and Degradation of
Organic Contaminants in Subsurface Environments, C. K. Smoley, Inc,
Chelsea, Michigan (1992).
9.
United States Environmental Protection Agency, Region IX, PRGs Table

2000 Update, San Francisco, California, November 2000.
10. American Petroleum Institute, APIs Decision Support System for

Exposure and Risk Assessments (APIDSS), Version 2.0, March 1999.
Available from API.
11. Spence, L.R., and British Petroleum, Risk-Integrated Software for Cleanups, (BP RISC) Users Manual Version 4.0 (2000). Available from
www.GroundwaterSoftware.com.
12. Boyd, S., and Sun, S., Residual Petroleum and Polychlorobiphenyl Oils as
Sorptive Phases for Organic Contaminants, in Soils. Envir. Science and
Tech., 24, 142-144 (1990).
109
Chapter 6
13. Johnson, P.C., and Ettinger, R.A., Heuristic Model for Predicting the
Intrusion Rate of Contaminant Vapors Into Buildings, Environmental
Science and Technology, 25, 1445-1452 (1991).
14. American Petroleum Institute, API Soil and Groundwater Research
Bulletin: Estimation of Infiltration and Recharge for Environmental Site
Assessment (1997). Available fo r download from www.api.org.
15. Guymon, G.L., Unsaturated Zone Hydrology, Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey (1994).
16. Howard, P.H., Handbook of Environmental Fate and Exposure Data For
Organic Chemicals (Volumes 1-3), Lewis Publishers, Chelsea, Michigan
(1989).
17. Howard, P.H., et al., Handbook of Environmental Degradation Rates, Lewis
Publishers Inc., Chelsea, Michigan (1991).
110
Chapter 7
Development of Reference Doses for Heavy

Total Petroleum Hydrocarbon (TPH)
Fractions
Deborah A. Edwards,1 Ann Tveit,1 and Michele Emerson2
1
2
ExxonMobil Biomedical Sciences, Inc.

Brown and Caldwell
Introduction
Over the last several years , a number of methods have been derived for
evaluating potential human health risks associated with exposures to petroleum
hydrocarbon mixtures. As described in previous chapters, the Total Petroleum
Hydrocarbon Criteria Working Group (TPHCWG) has conducted much of this
work, focusing primarily on refined petroleum products, such as gasoline and
diesel, which typically contain hydrocarbons below carbon number 35 (C35).
This chapter expands on the TPHCWGs work by taking into consideration
potential risks associated with exposures to crude oils and exploration and
production wastes which may contain heavy total petroleum hydrocarbon (TPH)
fractions (C35). The United States Environmental Protection Agency (USEPA)
and TPHCWG methodologies for developing reference doses (RfDs) and
reference concentrations (RfCs) are described. RfDs are recommended for
petroleum fractions C35 based on available toxicity studies.
USEPA Methodology for Determining RfDs and RfCs

In general, the USEPAs methods for determining RfDs and RfCs have
remained relatively unchanged since the release of Guidelines for Carcinogenic
Risk Assessment in 1986 [1]. Subsequent USEPA guidance which also discusses
the development of RfDs include: Risk Assessment Guidance for Superfund
Volume 1 Human Health Evaluation [2], Reference Dose (RfD): Description
and Use in Health Risk Assessments [3], Proposed Guidelines for Neurotoxicity
Risk Assessment [4], Guidelines for Reproductive Toxicity Risk Assessment [5],
and Guidelines for Developmental Toxicity Risk Assessment [6].
Definition of Reference Dose and Reference Concentration
Health risks associated with non-cancer effects, such as organ damage,
immunological effects, birth defects, and skin irritation, are assessed by
comparing an estimated average exposure to an RfD or an RfC. RfDs and RfCs
are not used to evaluate carcinogenic endpoints.
111
Chapter 7
The RfD for any given chemical is an estimate (with uncertainty spanning
perhaps an order of magnitude) of a daily exposure to the human population
(including sensitive subgroups) that is likely to be without appreciable risk of
deleterious effects during a lifetime [7]. Similar to the RfD, the RfC is also
used to assess non-cancer effects. The RfC is an estimate (with uncertainty
spanning perhaps an order of magnitude) of continuous inhalation exposure to
the human population (including sensitive subgroups) that is likely to be without
appreciable risk of deleterious effects during a lifetime [7]. Usually, exposures
that are less than the RfD or RfC are not likely to be associated with health risks
[2,8].
Derivation of RfDs and RfCs
Oral and dermal RfDs are calculated using the following equation:
RfD = NOAEL or LOAEL / (UF1 x UF2 x MF)
RfDs are typically expressed in units of milligram per kilogram per day
(mg/kg/day).
The RfD is derived from a critical dose or concentration level usually described
as a no-observed-adverse-effect level (NOAEL) or a lowest-observed-adverseeffect level (LOAEL) identified from toxicity studies in scientific literature. A
suitable uncertainty factor (UF) is applied to the NOAEL or LOAEL to reflect
differences between the study conditions and the human exposure to which the
reference value will be applied. In addition, a modifying factor (MF) may be
applied which reflects the completeness of the supporting scientific database.
Inhalation RfCs are derived from essentially the same equation as that used to
develop RfDs. However, the NOAEL is converted into units of milligram per
cubic meter (mg/m3) based on the exposure level. Thus, the units for inhalation
RfCs are expressed in mg/m3.
Selection of Appropriate Data for Development of RfDs and RfCs
The first step to develop an RfD or an RfC is to identify a critical study and
determine the NOAEL. The NOAEL is the highest dose at which no adverse
effects are observed. If the NOAEL is not available, then a LOAEL can be
used. However, use of the LOAEL adds an additional uncertainty factor into
development of the RfD or RfC. NOAELs and LOAELs are usually based on
laboratory dose-response experiments on animals that are exposed to relatively
high doses. To develop an oral RfD the most appropriate source of the NOAEL
or LOAEL is a chronic oral study. If there are no chronic oral data available,
then subchronic oral data can be used instead. Again this adds another level of
uncertainty into the derivation of the RfD. In some cases, chronic and
subchronic inhalation studies can be used to develop oral RfDs when no oral
112
Chapter 7
data are available. Most other toxicity data (i.e., dermal, acute, or genotoxic) are
not recommended for use in the development of RfDs. Oral studies are not used
in the development of inhalation RfCs.
Uncertainty and Modifying Factors
In general, UFs are applied in multiples of 10 and can range from 1 to 10,000.
Descriptions of some UFs are as follows:
Use a 10-fold factor when extrapolating from valid experimental results

in studies using prolonged exposure to average healthy humans. This
factor is intended to account for the variation in sensitivity among the
members of the human population and is referenced by USEPA as
10H.
Use an additional 10-fold factor when extrapolating from valid results

of long-term animal studies when results of human studies are either
not available or inadequate. This factor accounts for the uncertainty
involved in extrapolating from animal data to humans and is referenced
by USEPA as 10A.
Use an additional 10-fold factor when extrapolating from less than

chronic (lifetime exposure) results on experimental animals when there
are no useful long-term human data. This factor is intended to account
for the uncertainty involved in extrapolating from less than chronic
NOAELs to chronic NOAELs and is referenced by USEPA as 10S.
Use an additional 10-fold factor when deriving the RfD or RfC from a
LOAEL instead of a NOAEL. This factor is intended to account for the
uncertainty involved in extrapolating from LOAELs to NOAELs and is
referenced by USEPA as l0L.
The MF is an additional safety factor that is occasionally applied based on the

strength of the database and professional judgment. Typically, MFs range from
1 to 10 and are based on a subjective evaluation of the adequacy of the toxicity
data.
TPHCWG Methodology for Determining RfDs and RfCs

The general risk assessment methodology selected by the TPHCWG for
evaluating petroleum hydrocarbon mixtures is a combined indicator/surrogate
approach, which is consistent with USEPA methodology [1]. Indicators are
referred to as the single compounds within petroleum mixtures that are known to
be toxic. In the United States, they are typically evaluated and regulated
individually at either the federal or state level. Carcinogenic indicators are
113
Chapter 7
usually evaluated first because their presence (even in relatively low

concentrations) may drive a site cleanup, due to their greater relative toxicity. A
surrogate approach is used for the non-carcinogenic mixtures (petroleum
fractions) which represent the mass of petroleum remaining after evaluating the
indicators . In order to support this type of risk assessment approach, toxicity
criteria, such as RfDs and RfCs, must be developed for each of the petroleum
fractions.
The Massachusetts Department of Environmental Protection (MADEP) appears
to have developed the first fractional approach to evaluating the risk of TPH [9].
In 1995, British Columbia Environment (BCE) modified the MADEP approach
to include fate and transport of fractions and to be specific for ecological
receptors of concern in the province [10].
The TPHCWG evaluated subchronic, chronic, reproductive, developmental,
immunotoxicity, and neurotoxicity data on petroleum fractions available in
scientific literature. The USEPAs NOAEL-based methodology to develop oral
and dermal RfDs was followed, and route-to-route extrapolation was minimized,
i.e., oral studies were used for oral criteria, wherever possible. The development
of benchmark doses was considered as an alternative approach to using
NOAELs. However, the methodology for deriving a benchmark dose was
believed to be less established and more controversial [11].
The fraction-specific oral RfDs and inhalation RfCs recommended by the
TPHCWG are listed in Table 1. Petroleum fractions are based on environmental
fate and transport characteristics of the constituents within a given carbon range
rather than differences in toxicity. The methodology used to select the
equivalent carbon ranges is described in the TPHCWGs Volume 3 Selection of
Representative Total Petroleum Hydrocarbon (TPH) Fractions Based on Fate
and Transport Considerations [12].
114
Chapter 7
Table 1. Preliminary TPHCWG toxicology fraction-specific oral RfDs and

inhalation RfCs.
Aromatic Aromatic
Carbon Oral RfD Inhalation
Range 1 (mg/kg/Day) RfC (mg/m3) Critical Effect
Aliphatic 0.2
0.4
Hepatotoxicity
C5-C6
Nephrotoxicity
C>6-C8
Aromatic
C>7- C8
Aliphatic Aliphatic
Oral RfD Inhalation
5.0
18.4
Neurotoxicity
C>8-C10 0.04
(mg/kg/Day) RfC (mg/m3) Critical Effect
0.2
Decreased
Body Weight
0.1
1.0
Hepatic and
Hematological
Changes
NA
Nephrotoxicity 2.0
NA
Hepatic
(Foreign Body
Reaction)
Granuloma
C>10-C12
C>12-C16
C>16-C21 0.03 2
C>21-C35
1) Carbon range: equivalent carbon range [12].
2) This is the pyrene (C16) value.
RfD = Reference Dose.

RfC = Reference Concentration.
NA = Not available.
As shown in Table 1, the TPHCWG developed oral RfDs for aliphatic and
aromatic petroleum fractions ranging from C5-C35 and inhalation RfCs for the
aliphatic and aromatic fractions from C5-C16 [13]. Due to the limitations of the
experimental design of available toxicity studies, the TPHCWG did not develop
dermal RfDs for the petroleum fractions in these carbon ranges.
Based on both observed toxicity and uncertainty associated with the available
studies, the RfDs for the aromatic fractions are at least an order of magnitude
lower than those for the aliphatic fractions. Although more toxicity data were
available for the aliphatic fractions, the data available for aromatic fractions
indicate greater toxicity. In addition, greater uncertainty was factored into the
development of the aromatic fraction RfDs than into the aliphatic fraction RfDs
because toxicity data were limited on the aromatic fractions from C9-C16 and
from C17 -C35.
For the C17-C35 aromatic fraction, the TPHCWG assigned an oral RfD of
0.03 mg/kg-day based on the indicator compound pyrene with a carbon number
of 16 (C16). This was necessary because no other toxicity data were available
for any aromatic compound with a higher carbon number, with the exception of
carcinogenicity data on polycyclic aromatic hydrocarbons (PAHs). The
aromatic fraction C17-C35 contains PAHs that are normally considered to be
indicator compounds for risk assessment purposes .
115
Chapter 7
Toxicity Assessment of Petroleum Hydrocarbons Above C35

Petroleum hydrocarbons greater than C35 are not normally addressed from a risk
perspective in the United States. However, the toxicity of this heavier petroleum
fraction may need to be assessed for many types of crude oils or heavy refined
products (e.g., Bunker C). Therefore, the toxicity data for petroleum streams
with high boiling ranges were evaluated. A few of these toxicity studies were
useful for developing RfDs. These studies include two 28-day dermal toxicity
studies of two vacuum residuum samples and several 90-day oral studies on
white mineral oils and microcrystalline waxes (C22-C~80). No toxicity data were
available to assess the aromatic hydrocarbon fraction C35-C44. Therefore, the
RfD of pyrene (C16) must be used for this fraction, as the TPHCWG did for C17C35 aromatics.
White Mineral Oils and
Microcrystalline Waxes 90-Day Oral Studies
Toxicity data on aliphatic hydrocarbon fractions (high molecular weight white
mineral oils) were available for carbon numbers from C35 to approximately C44.
These white mineral oils decrease in toxicity by an order of magnitude as
compared to lighter molecular weight white mineral oils (C17-C~35.). The reason
for this decrease in toxicity was believed to be the reduced bioavailability of
these heavier compounds. Based on a 90-day oral study with white mineral oils,
an oral RfD of 20 mg/kg/day was developed for the aliphatic fraction containing
C35-C44 [13].
The same RfD for C35-C44 can be extended to include aliphatics up to C80 based
on 90-day oral studies on microcrystalline waxes [14]. These waxes are derived
from the highest boiling petroleum fractions and have viscosities ranging from
10-30 centistokes at 100C. These waxes are highly refined, saturated, aliphatic
hydrocarbon mixtures with <1% aromatics. Two microcrystalline waxes were
included in a 90-day oral toxicity study in Fischer 344 rats. One wax was a high
sulfur wax (HSW) containing an average carbon distribution of C20-C74. The
other was a high melting point wax (HMPW) with an average carbon
distribution of C22-C80. In this study, rats were fed diets containing either HSW
or HMPW at concentrations of 2, 20, 200, or 2,000 mg/kg/day. HSW and
HMPW did not produce adverse toxic effects as evidenced by organ weights,
hematology, clinical chemistry, or histopathology [14].
The NOAEL for this rat study is 2,000 mg/kg/day. Using an UF of ~100 (3 for
animal-to-human extrapolation, 10 for subchronic-to-chronic extrapolation, and
3 for individual susceptibility), an oral RfD of 20 mg/kg/day was developed.
This UF was also used by the TPHCWG for white mineral oils [13].
116
Chapter 7
In addition, the amount of hydrocarbons in the liver following oral exposure to

these HSW and HMPW was evaluated. No hydrocarbons were deposited in the
livers of animals fed microcrystalline waxes. Moreover, it was concluded that
no hydrocarbons with carbon numbers C35 (boiling point 491C) are passing the
intestinal membranes [15]. Therefore, applying an oral RfD of 20 mg/kg/day to
the C35 fraction is extremely conservative.
Vacuum Residuum - 28-Day Dermal Studies
The American Petroleum Institute [16,17] conducted two 28-day dermal studies
on vacuum residuum in rabbits. Vacuum residuum is defined as the residue
obtained from the vacuum distillation of crude oil after all other fractions have
been recovered, usually containing a mixture of hydrocarbon types >C44. The
vacuum residuum fraction of a crude oil is comprised of very large molecules
(those that boil above approximately 600C) that are not well characterized as to
their compositional make-up, but it is known to contain a mixture of aliphatics,
aromatics, metals, and asphaltenes. This fraction is also enriched in heteroatoms
(nitrogen, sulfur, and oxygen containing compounds) [18]. From a mobility
perspective, the high molecular weight hydrocarbons >C44 will not move
significantly from the area of release via groundwater. Therefore, they will
remain together as a mixture in the environment. In addition, these materials are
not volatile so exposure by inhalation would not be significant. Therefore, it is
appropriate to develop a RfD for the combined aromatic and aliphatic fraction
>C44. The risk posed by these materials would be limited to direct contact by
oral or dermal routes of exposure.
Rabbits were exposed to vacuum residuum at concentrations of 200, 1,000, and
2,000 mg/kg, three times a week for 4 weeks. Rabbits exposed to 2,000 mg/kg
displayed decreased food intake and skin lesions. When compared to controls,
no adverse systemic effects were noted in response to vacuum residuum
exposure. The NOAEL for these studies was 2,000 mg/kg. A preliminary
dermal RfD of 0.8 mg/kg/day was developed with the understanding that these
are only 28-day studies .
The preliminary dermal RfD of 0.8 mg/kg/day was developed by converting
2,000 mg/kg to units of mg/kg/day. To convert to units of mg/kg/day, the
mg/kg/week dose (2,000 mg/kg x 3 times a week) was divided by 7 to yield a
NOAEL of 800 mg/kg/day. In addition, an UF of 1,000 was used (a factor of 10
for each of the following: animal-to-human extrapolation, individual
susceptibility, and subchronic-to-chronic extrapolation).
These vacuum residuum studies do not lend themselves to development of an
oral RfD for >C44. Because it is generally accepted that oral bioavailability is
greater than dermal bioavailability, an oral RfD would probably be lower than
the derived dermal RfD. Based on the differences in bioavailability, it may be
117
Chapter 7
reasonable to assume an UF of 10 for dermal-to-oral bioavailability, suggesting

an oral RfD of 0.08 mg/kg/day could be developed for petroleum fractions C35.
This oral RfD is greater than the oral RfD of 0.03 mg/kg/day assigned by the
TPHCWG for aromatic fractions containing C17-C35.
Summary
Literature searches for toxicity data on heavy petroleum hydrocarbons (i.e.,
C35) were conducted and USEPA approved methods for developing RfDs were
followed. This work followed the same general procedures used previously by
the TPHCWG for developing petroleum fraction-specific RfDs. RfDs were
developed for separate aromatic and aliphatic fractions containing hydrocarbons
from C35 -C44. For the >C44 fraction, one RfD was developed for the combined
aliphatic and aromatic fraction, known as the vacuum residuum fraction.
For the aliphatic hydrocarbon fraction C35-C44, an oral RfD of 20 mg/kg/day was
developed from studies of high molecular weight white oils and microcrystalline
waxes. For the aromatic hydrocarbon fraction C35-C44, there were no data
available that could be used to develop either a dermal or an oral RfD. Lacking
any data for the aromatic hydrocarbon fraction C35-C44, an RfD of 0.03
mg/kg/day could be assigned, based on the RfD for pyrene, a C16 component of
petroleum. The TPHCWG used this approach for the C17-C35 aromatic fraction.
Based on negligible mobility, there is no scientific advantage to separating
aromatic from aliphatic hydrocarbon fractions above C44. A preliminary dermal
RfD of 0.8 mg/kg/day can be derived for a single hydrocarbon fraction (>C44).
This dermal RfD is based on dermal studies on vacuum residuum, which
contains aliphatic and aromatic compounds as well as resins and asphaltenes. It
is believed that using vacuum residuum as a surrogate for this fraction of crude
oil is conservative.
There were no oral toxicity studies found that could serve as the basis for an oral
RfD. It might be possible to develop an oral RfD based on the dermal data from
the vacuum residuum studies. However, the USEPA has not traditionally
accepted oral RfDs based on dermal data. If an oral RfD is developed based on
the dermal data from the vacuum residuum studies, then it may be reasonable to
assume an UF of 10 for dermal-to-oral bioavailability. This would suggest that
an oral RfD of 0.08 mg/kg/day could be developed for the single petroleum
fraction >C44. However, since this approach is not recommended by the
USEPA, the Petroleum Environmental Research Forum (PERF) 97-08 project
team chose to use a very conservative approach by applying an oral RfD of
0.03 mg/kg/day (based on the oral RfD of pyrene) and a dermal RfD of
0.8 mg/kg/day for the >C44 fraction.
118
Chapter 7
References
1.
United States Environmental Protection Agency (USEPA), Guidelines for

Carcinogen Risk Assessment, Federal Register, 51 (185), 33992-34003
(1986).
2.
United States Environmental Protection Agency (USEPA), Risk

Assessment Guidance for Superfund. Vol. 1: Human Health Evaluation
Manual, EPA/540/1-89/002, Washington, D.C. (1989).
3.
United States Environmental Protection Agency (USEPA), Reference

Dose (RfD): Description and Use in Health Risk Assessments,
Background Document 1A (March 15, 1993); Integrated Risk Information
System (IRIS) (1998).
4.
United States Environmental Protection Agency (USEPA), Proposed

Guidelines for Neurotoxicity Risk Assessment, Federal Register, 60 (192),
52032-52056 (1995).
5.

Reproductive Toxicity Risk Assessment, EPA/630/R-96/009, September
1996.
6.

Developmental Toxicity Risk Assessment, Federal Register, 51, 3402834040, October 1996.
7.
United States Environmental Protection Agency (USEPA), Integrated Risk

Information System. Glossary of IRIS Terms, revised October 1999.
8.
United States Environmental Protection Agency (USEPA), Health Effects

Assessment Summary Tables, FY 1997 Update, 9200.6-303 (97-1), EPA540-R-97-036, July 1997.
9.
Massachusetts Department of Environmental Protection (MADEP), Bureau

of Waste Site Cleanup and Office of Research and Standards, Background
Documentation for the Development of MCP Numerical Standards (1994).
10. British Columbia Environment (BCE), Recommendations to B.C.

Environment for Development of Remediation Criteria for Petroleum
Hydrocarbons in Soil & Groundwater, Vol. I and II, Victoria, B.C., June
1995.
11. United States Environmental Protection Agency (USEPA), The Use of the
Benchmark Dose Approach in Health Risk Assessment, EPA/630/R94/007 (1995).
12. Gustafson, S., Tell, S.G., and Orem, D., Selection of Representative Total
Petroleum Hydrocarbon (TPH) Fractions Based on Fate and Transport
Considerations, Total Petroleum Hydrocarbon Criteria Working Group
119
Chapter 7
Series, Vol. 3, Amherst Scientific Publishers, Amherst, Massachusetts

(1997).
13. Edwards, D.A., Andriot, M.D., Amoruso, M.A., Tummey, A.C., Bevan,
C.J., Tveit, A., and Hayes, L. A., Development of Fraction Specific
Reference Doses (RfDs) and Reference Concentrations (RfCs) for Total
Petroleum Hydrocarbons (TPH), Total Petroleum Hydrocarbon Criteria
Working Group Series, Vol. 4, Amherst Scientific Publishers, Amherst,
Massachusetts (1997).
14. Smith, S.H., Mallett, A.K., Priston R.A.S., Brantorn, P.G., Worrell, N.R.,
Sexsmith, C., and Simpson, B., Ninety-Day Feeding Study in Fischer 344
Rats of Highly Refined Petroleum Derived Food Grade White Oils and
Waxes, Tox. Path., 24, 214-230 (1996).
15. Freeman S., Simpson, B.S., and Tiet, P, White Oil and Waxes - Summary
of 90-Day Studies, CONCA WE Report 93/56, CONCAWE, Brussels
(1993).
16. American Petroleum Institute (API), 28-Day Dermal Toxicity Study in the
Rabbit: Vacuum Residuum (Sample 81-13), API Med. Res. Publ., 3032853 (1983).
17. American Petroleum Institute (API), 28-Day Dermal Toxicity Study in the
Rabbit: Vacuum Residuum (Sample 81-14), API Med. Res. Publ., 3032853 (1983).
18. Altgelt, K.H., and Boduszynski, M.M., Composition and Analysis of Heavy
Petroleum Fractions, Marcel Dekker (1994).
120
Chapter 8
Developing Total Petroleum Hydrocarbon

Risk-Based Screening Levels for Sites
Impacted by Crude Oils and Gas Condensates
Sara J. McMillen,1 Renae I. Magaw,1 Jill M. Kerr,2 and Lynn Spence3
1

3
2
Introduction
A review of upstream regulations summarized in Chapter 2 indicated that there
is a need to develop a risk-based approach for establishing total petroleum
hydrocarbon (TPH) regulatory limits, particularly for crude oils, condensates,
and their associated wastes at exploration and production (E&P) sites. To
accomplish this goal, a technically sound approach for determining the potential
risk of crude oils and condensates based on their composition must first be
developed, and then this risk-based approach to assessment and management
must be accepted by upstream environmental regulators. TPH risk-based
screening levels (RBSLs) address the risk of adverse non-cancer human health
effects associated with the bulk of the hydrocarbons in complex petroleum
mixtures. TPH RBSLs are first determined for each TPH equivalent carbon
fraction and then a single TPH RBSL is calculated based on the specific mixture
or oil composition, taking into account all of the individual TPH fractions. This
chapter will describe the TPH RBSLs developed for use in the United States
based on the composition of 70 crude oils and 14 condensates.
Developing TPH RBSLs

Procedures for deriving TPH RBSLs were developed by incorporating a
modified Total Petroleum Hydrocarbon Criteria Working Group (TPHCWG)
fractionation scheme, (see Chapters 4 and 5), fate and transport properties , (see
Chapter 4), and toxicity criteria developed for the equivalent carbon number
TPH fractions, (see Chapter 7), into the American Society for Testing and
Materials (ASTM) risk-based corrective action framework [1]. The potential
exposure scenario for which TPH RBSLs were developed was that of a spill of
crude oil or condensate at an oil or gas production site. Figure 1 illustrates the
exposure scenario.
121
Chapter 8
Impacted
soil
Volatilization
Leaching
Dissolution
Groundwater
Figure 1 . Potential exposure pathways for current production site workers .

Under current conditions at the site, workers may be exposed to contaminated
surface soil, or to vapors from subsurface soil or groundwater. In the future, the
land could be redeveloped for some other type of commercial or industrial use,
or it could be redeveloped for recreational or residential purposes. In these cases,
the future receptors could be exposed through the same exposure pathways that
currently apply to workers. Since most E&P sites are located in remote areas,
the most reasonably expected future land uses are generally non-residential, and
therefore the primary focus of the TPH RBSL development will be on the nonresidential case. However, it is possible that some sites will be redeveloped for
residential purposes, and therefore TPH RBSLs will also be derived for
residents .
Characteristics of Oils Used to Develop RBSLs
The 70 crude oils used to develop RBSLs were selected to cover a wide range of
American Petroleum Institute (API) gravity values and geographical locations.
The API gravity for these crude oils ranges from 8.8 to 46.4 and they are
representative of all crude oil types produced globally. More details on these
oils are presented in Chapter 4. Therefore, this data set can be used to set Tier 1
TPH RBSLs for all crude oils. The condensate data set is smaller (14 samples),
but it does represent a wide API gravity range (46 to 70).
RBSL Calculation Procedures
The RBSL calculations followed the general approaches and algorithms used by
the TPHCWG and in the ASTM Standard Guide for Risk-Based Corrective
Action applied at Petroleum Release Sites [1]. The variables and the specific
122
Chapter 8
values used to estimate exposure are presented in Tables 1 and 2. These are also
generally consistent with those presented in the ASTM standard; however, a few
modifications have been made to better reflect current risk assessment practices
in the United States . Recent scientific data and regulatory guidance suggest that
revision of some of the default values originally presented in the ASTM
standard is appropriate. Two variables, the relative dermal absorption factor
(RAFd) and the skin adherence factor (M), were updated to better reflect more
current state and United States Environmental Protection Agency (USEPA)
default values [2,3,4,5,6,7,8,9,10]. In the absence of chemical-specific
information, an RAFd of 0.1 and an M of 0.2 mg/cm2 were assumed for all
organic chemicals. These values are consistent with current United States
federal and state guidance.
Table 1. Summary of dermal absorption factors and skin adherence values used
by selected regulatory agencies.
Chemical
Organics
Polyaromatic Hydrocarbons
Volatile Organic Compounds

Semi- Volatile Organic Compounds
All Chemicals
Relative Dermal Absorption

Factor (RAFd)
0.10 [2,3,5,8]
0.10 (for Non-PAH) [4]
0.15 [4]
0.20 [9]
0.13 [2,3]
0.10 [6]
0.10 [10]
Soil to Skin Adherence (M), mg/cm2
0.2 [5,7,8,10]
1.0 [4]
0.08 for Adults and 0.3 for Children [2]
0.10 for Adults and 0.2 for Children [3]
Some previously published residential RBSLs (based on United States exposure

parameters) did not include potential exposures during childhood in the example
calculations [1,11]. Residential RBSLs should include the potentially higher
overall exposure levels that people may experience during childhood. Exposure
parameters for residential sites presented in Table 2 include childhood specific
ingestion and dermal exposure parameters. These parameters are consistent
with current risk assessment practice and with United States federal and state
regulatory guidance [8,12,13].
123
Chapter 8
Table 2. Exposure factors used in calculating RBSLs [8,10,12,13].
Definition
Ingestion Rate, Soil
Inhalation Rate (Outdoor)
Soil to Skin Adherence
Skin Surface Area
Body Weight
Exposure Time
Exposure Frequency
Exposure Duration
Averaging Time (Non-carcinogens)
Units
mg/day
m 3/day
mg/cm2
cm2
kg
hours/day
days/yr
years
years
Residential
Symbol Value
IRsoil
191
IRsoil
8.3
M
0.2
SA
2800
BW
15
ET
24
EF
350
ED
6
AT
6
NonResidential
Value
50
31
0.2
3300
70
8
250
25
25
The commercial exposure scenario, as defined in the ASTM guidance, was

reevaluated as part of this analysis and was found to be protective of other nonresidential exposure scenarios that may be more directly applicable to E&P
operations. For example, the ASTM commercial scenario was found to be
adequately protective for oil field workers and recreational land use. RBSLs
derived for adults in a commercial setting, as defined by ASTM, were equal to
or lower than those derived for adult oil field workers and adults and children in
a recreational setting. Based on interviews with E&P environmental health and
safety professionals, oil field workers were assumed to spend 30-40% of an
average workday in a truck traveling within or between E&P sites and 60-70%
of the day checking flows or volumes and performing maintenance activities.
The primary exposure route for these workers was assumed to be inhalation of
volatile materials and there was assumed to be little or no direct contact with
contaminated site soils, groundwater or surface water. Recreational receptors
were assumed to be on site for a full 2-week vacation period every year for
30 years or were assumed to be on site 4 hours per day, 2 days per week,
9 months per year for 30 years. Based on these findings, the commercial
exposure scenario is referred to as a non-residential (commercial and
recreational) scenario in this analysis.
A single, composite TPH RBSL was calculated for each of the 70 crude oils and
14 condensates. This composite RBSL was calculated by summing the Hazard
Quotients (HQs) for the 13 equivalent carbon number fractions. The Hazard
Quotient for any given fraction is the ratio of the level of exposure of that
fraction over a specified time period to the reference dose assigned to that
fraction. The sum of the individual Hazard Quotients is called the Hazard Index
(HI) and it is set equal to one. The HQ for each equivalent carbon number
fraction was calculated by taking the weight of the fraction multiplied by the
total TPH (mg/kg) and dividing by the RBSL of the fraction.
124
Chapter 8
RBSLS for Crude Oils and Selected Products

Residential and non-residential RBSLs were developed for 70 crude oils using
the parameters described above. Additionally, five products made from crude
oil (gasoline, baby oil, mineral oil, diesel, and Vaseline) were analyzed. The
modified TPHCWG fractions were determined for all these petroleum materials
and the data were used to calculate RBSLs. The modifications to the fractions
take into consideration those hydrocarbons that cover the entire range of
hydrocarbons that may be present in crude oil.
RBSLs were developed for both residential and non-residential land uses. The
following discussion focuses on the non-residential case, because it is the most
likely land use for E&P sites under current and reasonably expected future
conditions. Residential RBSLs are also presented; however, due to the overall
higher exposure potential at residential sites, and the potential for additional
exposure pathways to be present, it would generally be prudent to consider
collecting at least some site-specific information prior to applying Tier 1 RBSLs
to residential sites.
The RBSLs calculated for some products refined rfom crude oil (gasoline,
diesel, mineral oil, baby oil, and Vaseline) are shown in Table 3. All RBSLs
are in units of mg -TPH per kg-soil (mg/kg). The exposure pathway with the
lowest TPH RBSLs for all of the products involves direct human contact with
surface soil, as long as benzene is considered as a separate indicator compound
and is not included in the TPH RBSL. See Chapter 11 for an evaluation of
benzene risk.
The non-residential surface soil RBSL for baby oil is 269,000 mg/kg (or 26.9%),
for mineral oil is 313,000 mg/kg (or 31.3%), and for Vaseline the value is
333,000 mg/kg (or 33.3%). The calculated RBSLs for baby oil, mineral oil and
Vaseline validate that the assumptions used for the exposure pathways and for
the toxicity and fate parameters are extremely conservative, since these products
are known to be safe for human contact and/or ingestion. Baby oil, mineral oil
and Vaseline contain only high molecular weight aliphatic compounds and no
aromatic hydrocarbons (the more toxic and water soluble hydrocarbons), thus
they should have high RBSL values. In contrast, gasoline, which is composed
of hydrocarbons mainly ranging from C5 to C10 has the lowest RBSL. However,
benzene will drive the risk at gasoline spill sites, since gasoline products may
contain several percent benzene. Diesel which contains approximately 30%
aromatics, and has a molecular weight range of ~C10 to C22, has a RBSL of
42,000 mg/kg.
125
Chapter 8
Table 3. TPH risk-based screening levels for five refined petroleum products.
Gasoline
Diesel
Baby Oil
Mineral Oil
Vaseline
Leaching to
GW (mg/kg)
RES*
RES
RES
RES
RES
Non-Residential Scenario
Outdoor
Surficial
Vaporization (mg/kg) Soils (mg/kg)
RES
30,000
RES
42,000
RES
269,000
RES
313,000
RES
333,000
RES or residual saturation indicates that the oil does not pose a significant risk
to human health even when present as residual material in soil.
Non-residential TPH RBSLs for each of the 70 crude oils analyzed as part of
this study are plotted versus their API gravity in Figure 2. These RBSLs are
based on direct human contact with contaminated surface soils. Leaching to
groundwater and volatilization to outdoor air are not exposure pathways of
concern for crude oils because most of the equivalent carbon fractions found in
crude oils are either not soluble or volatile enough to cause a concern.
80,000
TPH Non-residential Surface Soil RBSL, mg/kg soil
70,000
60,000
50,000
40,000
30,000
20,000
10,000
0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
API Gravity
Figure 2. Non-residential surface soil TPH RBSLs calculated for 62 crude oils
from around the world plotted by their API gravity. (API gravity data are not
available for 8 of the 70 crude oils analyzed.)
126
Chapter 8
The non-residential surficial soil RBSLs are greater than or equal to 35,000 mgTPH/kg-soil as shown in Figure 2. The highest RBSLs were 67,300 and 67,900
mg/kg, which were obtained for two very waxy crude oils containing high
amounts of the larger molecular weight aliphatic hydrocarbons (these
hydrocarbons are the least toxic). While there does not seem to be a good
correlation between calculated RBSLs and API gravity, the data are plotted in
this manner to illustrate the range of RBSLs obtained and the diversity of crude
oil types analyzed.
RBSLs for Crude Oil Associated Wastes
Wastes from four different production areas plus oily soil or waste samples were
analyzed for their equivalent carbon fractions. The analytical results were then
used to calculate TPH RBSLs for each waste and the results are shown in
Table 4. The non-residential RBSLs for the associated wastes are compared to
those of the crude oils produced at the four sites.
Table 4. Non-residential RBSLs for crude oils and their associated wastes.
Leaching to
GW (mg/kg)
Outdoor
Vaporization
(mg/kg)
Non-Residential
Surficial Soils
(mg/kg)
RES*
RES
RES
RES
RES
RES
RES
RES
67,000
89,000
80,000
79,000
RES
RES
46,000
RES
RES
RES
RES
50,000
53,000
Field #3
Crude Oil
Oily Soil
RES
RES
RES
RES
48,000
59,000
Field #4
Crude Oil
Oily Soil
RES
RES
RES
RES
50,000
58,000
Field #1
Crude Oil
Tank Bottoms
Oily Soil
Oiled Road Material
Field #2
Crude Oil
Cyclone Separator
Sludge
Slop Oil
127
Chapter 8
Table 4 (Continued).
associated wastes.
Field #5
Tank Bottoms
Composted Tank
Bottoms
Non-residential RBSLs for crude oils and their
Leaching to
GW (mg/kg)
Outdoor
Vaporization
(mg/kg)
Non-Residential
Surficial Soils
(mg/kg)
RES
RES
RES
RES
56,000
52,000
RES or residual saturation indicates that the oil does not pose a significant risk
to human health even when present as residual material in soil.
For Field #1, the RBSLs for tank bottoms, soil from an old spill site (oily soil),
and oiled road material are shown in comparison to the crude oil RBSL. This oil
has an API gravity of 32.3 and is a waxy/paraffinic oil. The RBSLs for the
wastes are higher than the crude oil RBSL. For Field #2, the RBSLs for cyclone
separator sludge and slop oil are shown in comparison to that of the crude oil.
This crude oil is a heavy oil, which has an API gravity of 19.4. The RBSLs for
wastes at both Field #1 and #2 are higher than that of the parent crude oils,
largely due to the loss of volatile aromatic hydrocarbons from the wastes due to
natural weathering processes . Weathering reduces the amounts of volatiles and
increases the relative amount of waxes (high molecular weight aliphatics).
Oil impacted soil was collected at two different producing oilfields (Field #3,4)
in the United States . These samples were analyzed, along with crude oil from
each field, for their equivalent carbon number fractions. Table 4 lists the nonresidential RBSLs for each of these samples. The oily soil samples yield
consistently higher RBSLs compared to the parent crude oil from the field. As
with the wastes from Fields #1 and #2, the higher RBSLs for the oily soils are
due to the loss of volatile aromatic hydrocarbons from the soil. There is a
definite trend for E&P wastes (tank bottoms, oiled road material, and oily soils)
to have higher RBSLs than their parent crude oils. Therefore, the crude oil
RBSLs presented in this chapter should be conservative for E&P wastes
accidentally released or intentionally applied to soil (for example, landfarming).
RBSLs for Biodegraded Oily Wastes
The data from Field #5 (see Table 4) are for tank bottoms from a crude oil
storage tank, and the same material after composting to reduce the hydrocarbon
content via biodegradation. After composting, the RBSL was slightly lower,
52,000 mg/kg rather than 56,000 mg/kg TPH for the tank bottoms . This is
because the more recalcitrant hydrocarbons (the large molecular weight
aromatics) are present at higher concentrations as an overall percentage of the
128
Chapter 8
oil, and this tends to drive the RBSL lower. The tank bottoms initially contained
498,000 mg/kg TPH and after composting the TPH level was reduced to 11,000
mg/kg, so the RBSL could be easily met via biodegradation processes/
composting at this site.
RBSLs for Condensates
TPH RBSLs were developed for the 14 gas condensate samples. Figure 3
illustrates the results for non-residential surface soil exposure.
TPH Non-residential Surface Soil RBSL, mg/kg soil
140,000
120,000
100,000
80,000
60,000
40,000
20,000
0
0
10
20
30
40
50
60
70
80
API Gravity
Figure 3. Non-residential surface soil TPH RBSLs calculated for 14

condensates plotted by their API gravity.
The TPH RBSLs are greater than or equal to 39,600 mg-TPH/kg-soil. Most of
the condensates had non-residential surface soil RBSLs in the range of 40,000 to
60,000 mg/kg soil. The highest RBSL, 117,000 mg/kg soil, was obtained for a
very light condensate that contained no aromatics >C12. While there does not
seem to be a good correlation between calculated RBSLs and API gravity, the
data are plotted in this manner to illustrate the range of RBSLs obtained.
Condensates can contain fairly high concentrations of benzene and since
benzene is a carcinogen, it should be addressed separately at condensate spill
sites as described in Chapter 11.
Recommendations on Use of RBSLs in the United States
The ranges of TPH RBSLs for crude oils and condensates at residential and nonresidential sites are shown in Table 5. For both the residential and nonresidential scenario, the only pathway posing a significant risk to human health
is direct contact with contaminated soil (soil ingestion, inhalation of soil
129
Chapter 8
particles, and dermal contact). The 95th percentile values for the crude oil
RBSLs are also shown in Table 5. These 95th percentile values are
recommended for establishing generic TPH Tier 1 RBSLs for all crude oil types,
rather than the mean value or the minimum value. At individual E&P sites, use
of generic TPH measurements should be sufficient to demonstrate whether or
not Tier 1 levels have been attained. Since these 70 crude oils are representative
of the composition of all crude oils produced around the world, there is no need
to routinely analyze E&P site soils by the detailed fractionation analytical
technique.
Table 5. Residential and non-residential RBSLs for crude oils and condensates
(all data in mg/kg soil).
Residential Summary
Statistics (All Pathways)
Crudes
Condensates
4,200
7,200
2,800
5,000
Mean
95%
Range
Minimum 2,700
Maximum 6,300
5,000
18,000
Non-Residential Summary
Statistics (All Pathways)
Crudes
Condensates
48,300
52,500
41,300
39,600
35,000
67,300
39,100
117,000
Conclusions
TPH RBSLs have been calculated for crude oils based upon the specific
composition of each oil. Since the crude oils analyzed to date reflect the
composition of all crude oils produced around the world, there is no need to
routinely analyze E&P site soils by a detailed fractionation analytical technique.
Also, the RBSLs for E&P wastes were shown to be higher than the RBSL for
fresh crude oil within a given production field. Therefore, the TPH RBSLs
developed for crude oils are conservative for E&P associated wastes. The 95%
RBSLs are recommended for use at crude oil and gas condensate spill sites.
These RBSLs can be compared to TPH results obtained using simple,
inexpensive analytical methods that have been used historically at E&P sites
(these are described in Chapter 4 and include USEPA 418.1 and 8015).
There is no reason to perform the costly TPHCWG analytical fractionation
method at most E&P sites, unless there is a need to do a site specific analysis.
For example, site specific TPH fractionation data would be desirable if there is
reason to believe that the Tier 1 RBSLs are too conservative, or if the spill
occurs on residential property. This suggests that an E&P site can be adequately
managed using generic TPH analyses (e.g., EPA Method 8015, Texas 1005, or
other TPH measurements). These TPH results can then be compared to the
Tier 1 RBSL values for crude oil and gas condensates presented in Table 5.
130
Chapter 8
References
1.
ASTM, Standard Guide for Corrective Action Applied at Petroleum

Release Sites, ASTM E 1739-95, American Society for Testing and
Materials, West Conshohocken, Pennsylvania (1995).
2.
Louisiana Department of Environmental Quality Risk Evaluation/

Corrective Action Program (RECAP), adopted June 20, 2000.
3.

Program, 30TAC350 (1999).
4.
California Environmental Protection Agency, Preliminary Assessment

Guidance Manual, Cal/EPA Department of Toxic Substances Control
(1994).
5.
Colorado Department of Public Health and Environment, Soil Remediation

Objectives, A Policy Outlining the Process for Developing Site-Specific
Soil Remediation Objectives, December 1997.
6.
Michigan Department of Natural Resources, Environmental Response

Division Memorandum #8 Revision #4 Generic Residential Cleanup
Criteria (1995).
7.
Hawaii Department of Health, Risk-Based Corrective Action and Decision

Making at Sites With Contaminated Soil and Groundwater (1995).
8.
United States Environmental Protection Agency Region IX, Region 9

PRGs Table 2000 Update (2000).
9.
Massachusetts Department of Environmental Protection, Petroleum

Hydrocarbon Toxicity Project Report, Polycyclic Aromatic Hydrocarbon
Cancer Risk Evaluation Methods (1995).
10. United States Environmental Protection Agency, Region VI, Human Health
Medium-Specific Screening Levels, Dallas, Texas, September 2000.
11. McMillen, S.J., Magaw, R.I., Kerr, J.M., and Edwards, D.A., SPE Paper No.
52722, Risk-Based Screening Method for Evaluating Total Petroleum
Hydrocarbons at Upstream Sites, presented at the 1999 SPE/EPA
Exploration and Production Environmental Conference, Austin, Texas (1999).
12. United States Environmental Protection Agency, Office of Emergency and
Remedial Response, Risk Assessment Guidance for Superfund Volume I:
Human Health Evaluation Manual, September 29, 1989.
13. United States Environmental Protection Agency, Office of Research and
Development, Exposure Factors Handbook, EPA/600/P-95/002Fa,
August 1997.
131
Chapter 9
Hydrocarbon Transport From Oil and Soil to

Groundwater
Kirk T. O'Reilly,1 Renae I. Magaw,1 and William G. Rixey2
1
2

Department of Civil and Environmental Engineering, University of Houston
Introduction
The transport of hydrocarbons to groundwater is an important remediation
consideration for many petroleum release sites. There is often a concern that
hydrocarbons may leach from impacted soil, migrate to groundwater, and impact
the water source at unacceptable levels. Simple screening level fate and
transport models that do not account for Raoults Law suggest that this potential
exposure pathway may be of concern at many crude oil release sites.
Incorporating Raoults Law into the analysis indicates that crude oils generally
have a limited potential to impact groundwater at levels that would present
unacceptable human health risks.
Dissolved Hydrocarbons and Raoults Law

As described in Chapter 6, the maximum dissolved concentration of a chemical
in an organic mixture is a function of its pure compound solubility and the mole
fraction of the chemical in the organic phase. When a chemical is part of an
organic mixture, its solubility in water will be limited as follows:
Cw max = S eff = x * S
(1)
where:
Cw max = Maximum dissolved concentration of chemical in water (mg/L)
x
= Mole fraction of chemical in organic chemical mixture (mole
chemical/mole oil)
S
= Solubility of pure chemical in water (mg/L)
S eff = Effective solubility of chemical in water (mg/L)
Equation (1) is known as Raoults Law and it is commonly used to predict the
dissolved concentration of a chemical in water in contact with a hydrocarbon
mixture. Raoults Law is valid for compounds that are liquids at room
temperature. For solid compounds, such as the larger polycyclic aromatic
hydrocarbons (PAHs), the subcooled liquid solubility is used in place of the pure
chemical solubility (S) in Equation (1).
132
Chapter 9
When petroleum hydrocarbons (for example crude oil) are present as nonaqueous phase liquid (NAPL) in contact with groundwater, the dissolved
concentrations of the individual chemicals are generally not affected by the
amount of oil in contact with the water as illustrated below1:
Mass of crude oil = mass of water:
Crude
1,000 mg/kg Benzene
Water
1.8 mg/L Benzene
Mass of crude oil < mass of water:

Water
1.8 mg/L Benzene
Crude: 1,000 mg/kg Benzene
Mass of crude oil > mass of water:
Crude
1,000 mg/kg Benzene
Water: 1.8 mg/L Benzene
Since the effective solubility of any individual chemical depends on its
concentration in the oil, but not on the amount of oil present [Equation (1)],
compositional information for an oil can be used to determine if a given
compound in oil, or the oil itself, could present a risk to groundwater. Raoults
Law can be used to predict the effective solubilities (i.e., the maximum
dissolved concentrations) of compounds of interest.
The predicted
concentrations can then be compared to groundwater protection goals, and if the
predicted values are less than the protection goals, then the oil cannot pose an
unacceptable ris k to human health via the groundwater ingestion pathway.
Only when the mass of the crude oil is very low does the amount of the crude
oil present affect the effective solubility of individual compounds. At low oil
concentrations, lower final concentrations in both phases can be expected due to
partitioning into the soil organic matter as illustrated in Figure 1.
133
Chapter 9
Crude Oil Impacts on Groundwater

PAH and Benzene Impacts on Groundwater
In Table 1, Equation (1) and data from Chapters 10 and 11 were used to
calculate maximum dissolved concentrations of aromatic compounds in water in
contact with a crude oil. These values were then compared to United States
groundwater protection regulations, in the form of United States Environmental
Protection Agency (USEPA) Maximum Contaminant Levels (MCLs ) for
drinking water [1]. For chemicals that do not have MCLs, alternate health-based
limits in the form of USEPA Region IX Preliminary Remediation Goals (PRGs)
for tap water were used [2]. The PRGs are analogous to MCLs and are
developed by USEPA using similar assumptions. Kerr et al. (2000) analyzed 60
oils for several aromatic compounds [3]. The highest concentration of each of
four representative compounds (Co max. in the table) was used in this example
and the molecular weight of the oil was assumed to be 200 g/mole. Since the
highest concentration of each compound in any oil was used, this is a worst case
evaluation.
Table 1. Calculated maximum dissolved concentrations, Cw max., of selected
aromatic compounds in crude oil.
Benzene
Naphthalene
Chrysene
Benzo(a)pyrene
Co max.
(mg/kg )1
5,900
3,700
120
7.7
X max. 2
1.51E-02
5.77E-03
1.05E-02
6.11E-06
S3
(mg/L)
1.80E+03
1.03E+02
1.65E-01
4.00E-02
Cw max.
(mg/L)4
2.72E+01
5.94E-01
1.74E-05
2.4E-07
MCL 5
(mg/L)
5.0E-03
NA
NA
2.0E-04
PRG6
(mg/L)
3.5E-04
6.2E-03
9.2E-03
9.2E-06
>MCL or
PRG
Yes
Yes
No
No
1) C o max = maximum concentration in crude oil (from measurements of 60 crude oils).

2) Xmax = maximum mole fraction (from measurements of 60 crude oils).
3) S = pure compound solubility. Values are from NMED Guidelines Table 4-3 [4]. Subcooled
liquid solubility if solid at 25C; calculated by multiplying the liquid solubility by the liquidsolid fugacity ratios from Peters et al. [5].
4) C w max = maximum dissolved concentration, calculated from Equation (1).
5) Values for benzene and benzo(a)pyrene are USEPA MCLs (there are no USEPA MCLs for
naphthalene and chrysene) [1].
6) USEPA Region IX Preliminary Remediation Goal (PRG) for tap water [2].
If any of the 60 crude oils were to be in direct contact with groundwater,

benzene and naphthalene would be the only compounds that could possibly have
dissolved concentrations that would exceed health-based limits. This same
analysis was extended to 13 PAH compounds for which sufficient data were
available. Even at the maximum concentrations for the 60 oils analyzed, the
maximum dissolved concentrations for all PAHs, other than naphthalene, are
below health-bas ed limits as shown in Table 2.
134
Chapter 9
Table 2. Calculated maximum dissolved concentrations of PAHs in crude oils.
PAH
Naphthalene
Acenaphthene
Anthracene
Fluorene
Benzo[a]anthracene
Fluoranthene
Chrysene
Pyrene
Dibenz[a,h]anthracene
Benzo[a]pyrene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Indeno[1,2,3cd]pyrene
Cw
Co max.
S
max.
1 X max.2
3
(mg/L)4
(mg/kg )
(mg/L)
3,700
5.77E-03 1.03E+02 5.9E-01
MCL 5
(mg/L)
NA
PRG6
(mg/L)
6.2E-03
>MCL
or PRG
Yes
58
17
380
38
26
120
82
9.2
7.7
14
7
1.7
NA
NA
NA
NA
NA
NA
NA
NA
2.0E-04
NA
NA
NA
3.7E-01
1.8E+00
2.4E-01
9.2E-05
1.5E+00
9.2E-03
1.8E-01
9.2E-06
9.2E-06
9.2E-05
9.2E-04
9.2E-05
No
No
No
No
No
No
No
No
No
No
No
No
7.52E-05
1.91E-05
4.57E-04
3.33E-05
2.57E-05
1.05E-04
8.1E-05
6.60E-06
6.11E-06
1.11E-05
5.55E-06
1.23E-06
2.12E+01
4.43E+00
1.24E+01
2.35E-01
1.08E+00
1.65E-01
1.04E+00
6.23E-01
4.00E-02
3.85E-02
6.15E-02
5.50E-04
1.6E-03
8.5E-05
5.7E-03
8.0E-06
2.8E-05
1.7E-05
8.4E-05
4.0E-06
2.4E-07
4.2E-07
3.4E-07
6.8E-10
1) C o max = maximum concentration in crude oil (from measurements of 60 crude oils).

2) Xmax = maximum mole fraction (from measurements of 60 crude oils).
3) S = pure compound solubility. Values are from NMED Guidelines Table 4-3 [4]. Subcooled
liquid solubility if solid at 25C; calculated by multiplying the liquid solubility by the liquidsolid fugacity ratios from Peters et al. [5].
4) C w max = maximum dissolved concentration, calculated from Equation (1).
5) Value for benzo(a)pyrene is the USEPA MCL (there are no USEPA MCLs for the other PAHs)
[1].
6) USEPA Region IX Preliminary Remediation Goal (PRG) for tap water [2].
TPH Impacts on Groundwater

TPH fractionation schemes are beginning to be used by state regulatory agencies
within the United States to assess the potential migration and risks associated
with petroleum hydrocarbons [6-10]. Raoults Law can be used to evaluate the
potential impacts of TPH on groundwater in the same way that it can be used to
evaluate the potential impacts of benzene and PAHs. A similar exercise to that
presented above for benzene and PAHs can be performed for TPH fractions.
However, because there are no USEPA MCLs or PRGs for TPH or TPH
fractions, the groundwater Protective Concentration Levels (PCLs) set by the
Texas Natural Resource Conservation Commission (TNRCC) as groundwater
protection goals are used for comparison [8]. In the TNRCC program, screening
level (Tier 1) health-risk based PCLs are available for 12 TPH fractions for the
ingestion of groundwater (GWGW Ing) exposure pathway.
By rearranging Equation (1), and substituting GWGW Ing (multiplied by a dilution
factor [8]) for Cw max, we can calculate the maximum concentration in the oil
phase for each TPH fraction that would prevent TPH levels in groundwater from
exceeding protective levels.
135
Chapter 9
GW
X=
GWIng * LDF
S
(2)
where:
GW
GW Ing = PCL for ingestion of groundwater (mg/L). See Table 3 of [8].

LDF
= Lateral dilution factor for groundwater (dimensionless) from Figure:
30 TAC 350.75 (b) of [8].
In Equation (2), the quantity GWGW Ing *LDF is equal to the effective solubility
[Cw max in Equation (1)] and represents the water concentration in direct contact
with soil and residual oil. The terms GWGW Ing and LDF are used here to be
consistent with notation used in the TNRCC rules.
Table 3 presents groundwater protection levels for various aromatic and
aliphatic fractions, their respective pure solubilities, and the calculated
maximum acceptable oil mole fractions for these fractions. These data indicate
that the levels for the various aromatic fractions in crude oil can be greater than
9 mole %. Note that for all aliphatic fractions with carbon numbers greater than
6 and for aromatic fractions with carbon numbers greater than 12, even if these
fractions existed as a pure liquid in soil, they could not pose an unacceptable
risk to human health based on the groundwater ingestion pathway.
Table 3. Calculated mole fractions in oil for various TPH fractions.
GW
>C7-C8 Aromatics (TPH)

>C8-C1 0 Aromatics (TPH)
>C21-C44 Aromatics (TPH) 3
C6 Aliphatics (TPH)
>C6-C8 Aliphatics (TPH)
>C8-C1 0 Aliphatics (TPH)
>C16-C44 Aliphatics (TPH) 3
1)
2)
3)
GW Ing 1
(mg/L)
2.4
0.98
0.98
0.98
0.73
0.73
1.5
1.5
2.4
2.4
2.4
49
S2
(mg/L)
530
65
25
5.8
0.65
0.0066
36
5.4
0.43
0.034
0.00076
2.5E-06
Concentration in
TPH (Mole Fraction)
0.09
0.30
0.78
1
1
1
0.83
1
1
1
1
1
From Table 3 of Texas Risk Reduction Program Rules (TRRP). Corresponding to a

residential land use scenario. LDF = 20.
From Figure: 30 TAC 350.73 (e) of TRRP Rules.
Extended for the C35 to C44 equivalent carbon number range for crude oils. Values
listed are from TRRP Rules for up to C35.
136
Chapter 9
Another interpretation of the data in Table 3 is that if the concentrations of these

fractions in a crude oil (or petroleum hydrocarbon mixtures in general) are
below those shown in Table 3, then no amount of TPH in the soil could produce
groundwater concentrations that would exceed protective levels. In this case,
acceptable TPH levels in soil would be limited by other exposure pathways,
rather than by leaching to groundwater. In general, this technique could be used
as a screening method for evaluating the potential of any given oil to produce
groundwater TPH levels that would exceed groundwater protective levels.
Impact of Soil Organic Content on Partitioning

When two organic phases and water are in contact, a chemical will partition
between the three phases. The equilibrium concentration in each phase will
depend on the relative affinity of the chemical for the three phases. In the
following example, the affinity of the chemical is 100 times greater for Phase 1
than water, and 75 times greater for Phase 2 than water. This is similar to the
relative affinity of petroleum hydrocarbons for oil and soil organic matter,
respectively. At equilibrium, the ratio of the concentration of a chemical in the
three phases remains constant at 100:75:1.
Organic Phase 1
100 mg/kg
Organic Phase 2
75 mg/kg
Water
1 mg/L
In unimpacted soil, both the soil organic content (SOC) and water are free of
petroleum hydrocarbons. When oil is added to soil, an individual chemical
contained in the oil will partition between the oil, SOC, and water. At low oil
levels, the concentration of the chemical in the oil phase decreases due to loss to
the SOC and water. This results in a lower dissolved concentration than would
be expected for oil-water partitioning alone. As the mass of oil increases
relative to the SOC, the final organic phase concentration remains closer to the
initial oil concentration. The dissolved concentration gets closer to, but cannot
exceed, the effective solubility predicted by Raoults Law.
The partitioning between oil, SOC, and water is demonstrated in the following
example. The concentration of the chemical of interest is initially 1,000 mg/kg
in the oil phase. Raoults Law predicts a maximum dissolved concentration of
10 mg/L. The amount of oil ranges from 100 to 10,000 mg/kg soil. The SOC is
1,000 mg/kg soil. Co f is the final concentration of the compound in the oil
phase. Cs is the concentration in the SOC phase and Cw is the dissolved phase.
137
Chapter 9
The dissolved concentration never exceeds the 10 mg/L effective solubility

predicted by Raoults Law.
Table 4. Partitioning behavior of a chemical between oil, SOC, and water.
Amount
of Oil
(mg/kg)
100
1,000
10,000
Amount
of SOC
(mg/kg)
1,000
1,000
1,000
Final Conc. in
Oil Phase (Cof
mg/kg)
118
571
930
Conc. of Chemical
in SOC Phase
(C s mg/kg)
88
429
698
Conc. of
Chemical in
Dissolved Phase
(Cw mg/L)
1.18
5.71
9.30
Figure 1 shows the same effect over a wider range of concentrations. The
dissolved concentration of the compound increases with oil concentrations at
low levels, but the maximum dissolved concentration remains limited to the
value predicted by Raoults Law.
100.00
Cw (mg/L)
10.00
Cw
Effective
Solubility
Predicted
by
Raoult's
Law
1.00
0.10
10
100
1000
10000
100000
1000000
Oil in Soil (mg/kg)
Figure 1. The partitioning behavior of a chemical between dissolved and soil

phases.
The partitioning of organic compounds between the SOC and water is the basis
for setting soil screening levels that are protective of groundwater. The USEPA
Soil Screening Guidance presents the following equation for relating soil
concentrations to dissolved concentrations [11]:
138
Chapter 9
Cs = Cw * [Kd + ((T w + T a KH )/?b))]
(3)
where:
Cs = Soil screening level (mg/kg)
Cw = Dissolved groundwater protection goal (mg/L)
Kd = Soil-water sorption coefficient (mg/kg dry soil)/(mg/
L water)
T w = Water filled porosity (cm3 water/cm3 soil)
T a = Air filled porosity (cm3 air/cm3 soil)
KH = Henrys Law coefficient (mg/L vapor/mg/L water)
?b = Soil bulk density (g dry soil/cm3 soil)
The component of the equation (T w + Ta KH )/?b takes into account the
fraction of the COCs in the water and vapor phase. Since this term is typically
much smaller than Kd, Equation (3) can be simplified for this discussion to:
Cs = Cw * Kd
(4)
The soil / water sorption coefficient Kd is a function of the amount of organic

phase associated with the soil, and the K for that organic phase. For a soil in
the absence of a NAPL or oil phase [12]:
Kd = Koc * foc
(5)
where:
Koc = SOC / water partition coefficient (mg /kg SOC)/(mg /L water)
foc = The fraction of organic carbon in the soil (kg SOC/kg soil)
If two organic phases are present, the organic carbon in the soil and the oil, then
the soil-water sorption coefficient in Equation (4) can be redefined as:
Kd = (Koc1 * foc1) + (Koc 2 * foc2)
(6)
An oil impacted soil has two organic phases, the SOC and the oil. By replacing
Koc2 with Koil (mg/kg oil)/(mg/L water), and foc 2 with the oil concentration in
the soil (kg oil/kg soil), Kd from Equation (6) can be inserted into Equation (4):
Cs = Cw * ((Koc * foc) + (Koil * foil))
(7)
Koil can be calculated using Raoults Law:
139
Chapter 9
Koil =
MWi
10 6 mg / k g
S i MWo
(8)
where:
MW i = Molecular weight of the compound (g/mole)
MW o = Average molecular weight of the TPH (g/mole)
Si
= Solubility of the compound of interest (mg/L)
Summary
The protection of groundwater is a critical component of environmental
management at sites impacted with petroleum hydrocarbons. The potential for
chemicals of concern to leach from oil or petroleum impacted soil must be
understood in order to accurately assess the risk to human health from the
ingestion of groundwater. In this chapter, the technical basis for calculating
maximum dissolved hydrocarbon concentrations in leachate was presented. If
the concentration of a constituent in either the oil or soil phase is known, the
maximum dissolved concentration can be calculated.
The dissolved
concentration is limited by equilibrium partitioning between the oil and water
phases. Partitioning between soil organic matter and these phases can further
reduce the dissolved hydrocarbon leachate concentration. The analysis of crude
oil data presented in this chapter illustrates that benzene and naphthalene may
potentially leach from crude oil-containing soil to groundwater at levels that
may pose an unacceptable risk to human health. The other priority pollutant
PAHs are not present in crude oil at high enough concentrations to be a potential
threat to groundwater via leaching from soil.
References
1.
EPA, Drinking Water Regulations and Health Advisories, Office of

Water, Washington, D.C. (1995).
2.
EPA Region IX, PRGs Table 2000 Update, San Francisco, California,
November 3, 2000.
3.
Kerr, J.M., Hamed, M.M., Melton, H.R., Magaw, R.I., McMillen, S.J., and
Naughton, G., Risk-Based Soil Screening Levels for Crude Oil: Role of
Polyaromatic Hydrocarbons, in the Proceedings of the 6th International
Petroleum Environmental Conference, Houston, Texas, K.L. Sublette, ed.,
SCG, Inc., Tulsa, Oklahoma, 474-488 (2000).
4.
New Mexico Environmental Department (NMED), Guidelines for

Corrective Action, New Mexico Underground Storage Tank Bureau,
March 2000.
140
Chapter 9
5.
Peters, C.A., Mukherji, S., Knightes, C.D., and Weber, W.J., Phase
Stability of Multi-Component NAPLs Containing PAHs, Environmental
Science and Technology, 31, 2540-2546 (1997).
6.
Alaska Department of Environmental Conservation (ADEC), Oil and

Hazardous Substances Regulations (1999).
7.
Massachusetts Department of Environmental Protection (MADEP),

Interim Final Petroleum Report: Development of Health-Based
Alternatives to the Total Petroleum Hydrocarbon (TPH) Parameter (1994).
8.
Texas Natural Resource Conservation Commission, Subchapter D Development of Protective Concentration Levels, in Texas Risk Reduction
Program (TRRP) Rule, 30 TAC 350 (1999).
9.
Texas Natural Resource Conservation Commission, Development of

Human Health PCLs for Total Petroleum Hydrocarbon Mixtures, RG366/TRRP-27 (2000).
10. Synowiec, K, Salhotra, A.M., and Varghese, B.K., The Application of

TPH With the Risk Based Decision Making Framework - A Nationwide
Survey, in Proceedings Petroleum Hydrocarbons and Organic Chemicals
in Groundwater: Prevention, Detection and Remediation, NGWA, 271-282
(1999).
11. U.S. Environmental Protection Agency, Soil Screening Guidance:
Technical Background Document (TBD), EPA/540/R-95/128 (1996b)
12. Boyd, S., and Sun, S., Residual Petroleum and Polychlorobiphenyl Oils as
Sorptive Phases for Organic Contaminants in Soils, Environmental Science
and Technology, 24, 142-144. (1990).
141
Chapter 10
Risk-Based Soil Screening Levels for Crude Oil:

The Role of Polyaromatic Hydrocarbons
Jill M. Kerr,1 Sara. J. McMillen,2 Renae I. Magaw,2 H. Rodger Melton,1 and
George Naughton 3
1

3
Arthur D. Little
2
Introduction
Polyaromatic hydrocarbons (PAHs) are ubiquitous in the environment and result
from both naturally-occurring and man-made sources. PAHs can be formed
whenever organic substances are exposed to high temperatures. Some PAHs are
also synthesized by several species of plants and bacteria [1]. The largest single
source of PAHs to the environment is likely the residential burning of wood [2].
Combustion of fossil fuels is also a major source, with other sources including
volcanoes , agricultural burning, asphalt roads, and forest fires [2]. In the home,
PAHs can be found in tobacco smoke, grilled and smoked foods, wood-burning
fireplaces, meat, processed or pickled foods, and beverages.
Most of the direct releases of PAHs are to the atmosphere, and most of these are
associated with particulate matter. Settling of the particulates allows PAHs to
be redistributed to other environmental media, such as soil and water. PAHs are
found in relatively high concentrations at certain types of industrial sites,
particularly in contaminated soils at manufactured gas plants (MGP) and woodtreatment facilities. Creosote, a common wood preservative, is a distillate of
coal tar (distilled from coal at high temperatures in the absence of oxygen), and
contains higher concentrations of PAHs than many other substances [3].
At sufficient dose levels, laboratory studies show that some PAHs cause adverse
health effects including cancer and reproductive difficulties in animals. People
exposed for long periods of time to mixtures of chemicals that include PAHs can
also develop cancer [2]. Therefore, PAHs have been identified as chemicals of
potential environmental concern, and PAHs known to cause adverse effects have
been placed on the United States Environmental Protection Agency (USEPA)
priority pollutant list.
The structures of the 16 PAHs currently on the USEPA priority pollutant list are
shown in Figure 1. Selected physical-chemical properties for these PAHs are
given in Table 1. Boiling points range from 217 to 542C and aqueous
solubilities of pure solids range from 0.00026 to 31 mg/L.
142
Chapter 10
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Anthracene
Phenanthrene
Fluoranthene
Pyrene
Benz [a]anthracene
Chrysene
Benzo [b]fluoranthene
Benzo [k]fluoranthene
Benzo [a]pyrene
Indeno [1,2,3-cd]
pyrene
Dibenz [a,h]anthracene
Benzo [g,h,i]perylene
Figure 1. Chemical structures of the 16 USEPA priority pollutant PAHs.

Table 1. Names and selected physical-chemical properties of the 16 USEPA
priority pollutant PAHs [4,5,6].
PAH
Naphthalene
Acenaphthene
Acenaphthylene
Anthracene
Phenanthrene
Fluorene
Fluoranthene
Benz[a]anthracene
Chrysene
Pyrene
Benzo[a]pyrene
Benzo[g,h,i]perylene
Indeno[1,2,3 -cd]-pyrene
1)
Boiling
Point 1 (C)
217
279
265
340
340
293
295
400
448
360
496
481
480
535
542
530
Aqueous
Solubility of Pure
Solid (mg/L)
31
3.8
16.1
0.045
1.1
1.9
0.26
0.011
0.0015
0.132
0.0038
0.0015
0.0008
0.0005
0.00026
0.062
Aqueous Solubility
of Pure Subcooled
Liquid (mg/L)
103.23
19
73.094
4.59
4.07
11.875
5.26
0.275
0.1545
1.015
0.1265
0.0384
0.0615
0.125
0.0866
1.55
Kp1
log
3.37
4.33
4.07
4.45
4.46
4.18
5.33
5.61
5.61
5.32
6.04
6.57
6.84
5.97
7.23
7.66
Vapor
Pressure
(Torr at 20C)
4.92 x 10-2
2.0 x 10-2
2.9 x 10-2
1.96 x 10-4
6.80 x 10-4
1.3 x 10-2
6.0 x 10-6
5.0 x 10-9
6.3 x 10-7
6.85 x 10-7
5.0 x 10-7
5.0 x 10-7
5.0 x 10-7
1.0 x 10-10
1.0 x 10-10
1.0 x 10-10
LogKp = Logarithm of the soil:water partition coefficient.
PAHs occur naturally in crude oils. While crude oils and their associated wastes
contain PAHs, there are few published data on the amounts of priority pollutant
143
Chapter 10
PAHs in exploration and production (E&P) wastes. The American Petroleum

Institute (API) analyzed crude oil-containing soils and tank bottoms for
chrysene, naphthalene, phenanthrene, and 1-methyl naphthalene [7]. In 1994,
the USEPA Office of Solid Waste sampled crude oil tank bottom wastes and
analyzed them for semi-volatile organics including naphthalene, phenanthrene,
fluorene, and chrysene [8]. Neither of these studies examined the entire suite of
priority pollutant PAHs in E&P wastes. However, their results indicate that the
distribution pattern and concentrations of PAHs in crude oils probably differ in
many respects from those found in soils containing PAHs from MGP and woodtreating facilities.
Except for the State of California, state regulatory agencies in the United States
do not require routine analysis of PAHs in wastes or soils at E&P sites.
However, at some E&P sites, PAHs have been included in site characterization
and risk assessment activities. It is important to understand the types and
concentrations of priority pollutant PAHs in crude oils and their associated
wastes in risk-based decision-making (RBDM) for E&P sites.
Analytical Methodology
Sixty crude oils from production sites around the world were analyzed for the 16
priority pollutant PAHs. Figure 2 illustrates the sources of the crude oils, by
geographic region. The crude oils were selected based on their diversity in API
gravities and compound classes. The goal was to analyze a diverse group of oils
that would represent the range of crude oil types produced around the world. In
addition to the 60 crude oils, 10 condensates, 10 oily wastes (tank bottoms,
sludges, etc.), and 6 oil-containing soils also were analyzed for their PAH
content.
144
Chapter 10
10
9
3 1
1
8
2
53
3
33
3
region from which they originated.
Using pentane, a dilution was prepared for each crude oil and condensate to
obtain an approximate concentration of 5 mg/mL. Each dilution was then
spiked directly with the following surrogates; naphthalene-d8, acenaphthene-d10,
phenanthrene-d10, and benzo[a]pyrene-d12. Extracts were obtained from wastes
and soils using either methylene chloride or pentane as the solvent. The diluted
crude oils and extracts were then passed through a silica gel chromatography
column to separate the aromatic hydrocarbons using modified USEPA Method
3630.
Pentane was used to elute the saturate fraction, methylene
chloride:pentane (50:50) was used to elute the aromatic hydrocarbons. The
aromatic fraction was concentrated and spiked with the following internal
standards; chrysene-d12 and fluorene d10.
To determine the concentrations of selected PAHs, the concentrated extracts and
diluted crude oils were analyzed by gas chromatography/mass spectrometry
(GC/MS) operated in the selected ion monitoring (SIM) mode using a modified
USEPA Method 8270. A five-point calibration, an Instrumental Reference
Material (NIST IRM 1491), and an oil reference standard (Alaska North Slope
crude oil) were analyzed at the beginning of each instrumental sequence. All
instruments were calibrated with analytical standards prior to the analysis of the
samples. Target analyte concentrations of the 16 priority pollutant PAHs were
calculated versus the internal standard compound and were corrected for
recovery efficiency of the surrogate compounds. The recovery of the surrogate
compounds was calculated relative to the internal standards added to the
samples prior to instrumental analysis.
145
Chapter 10
PAH Content of Analyzed Samples

The range and mean concentration of each of the priority pollutant PAHs as well
as the detection frequency for crude oils and condensates are provided in
Tables 2a and 2b. Naphthalene, phenanthrene, fluorene, chrysene, pyrene, and
benzo[b]fluoranthene occurred in >97% of the crude oils tested. Anthracene,
fluoranthene, dibenz[a,h]anthracene, and indeno[1,2,3-cd]pyrene were less
frequently detected (40% of oils contained anthracene and fluoranthene, and
<7% contained indeno[1,2,3-cd]pyrene). Acenaphthylene was not detected in
any of the 60 crude oils.
Table 2a. PAH content of 60 crude oils (all data are reported in mg/kg oil).
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Anthracene
Phenanthrene
Fluorene
Benz[a]anthracene
Fluoranthene
Chrysene
Pyrene
Benzo[a]pyrene
Indeno[1,2,3-cd]pyrene
Minimum
(mg/kg)
1.2
ND
ND
ND
ND
1.4
ND
ND
4
ND
ND
ND
ND
ND
ND
ND
Maximum
(mg/kg)
3700
NA
58
17
916
380
38
26
120
82
9.2
7.7
14
7
1.7
9.6
Mean
(mg/kg)
422.9
NA
13.9
3.4
176.7
73.6
5.5
3.9
28.5
15.5
1.0
2.0
3.9
0.46
0.06
1.53
Detection
Frequency
(%)
100
0
80
40
98
100
67
40
100
97
47
75
100
93
7
63
146
Chapter 10
Table 2b. PAH content of 10 condensates (all data are reported in mg/kg
condensate).
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Anthracene
Phenanthrene
Fluorene
Benz[a]anthracene
Fluoranthene
Chrysene
Pyrene
Benzo[a]pyrene
Benzo[g,h ,i]perylene
Minimum
(mg/kg)
200
ND
ND
ND
ND
3.9
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Maximum
(mg/kg)
5,700
9.2
12
27
250
82
0.78
11
5.5
12
NA
NA
2
NA
NA
NA
Mean
(mg/kg)
1,690
1.15
1.43
5.91
90
44.8
0.30
2.47
1.93
2.96
NA
NA
0.64
NA
NA
NA
Detection
Frequency
(%)
100
10
10
50
90
100
10
30
40
40
0
0
30
0
0
0
PAHs in bold italic have been shown to be carcinogenic in laboratory animals

[9].
ND = Below detection limits. Detection limit = 0.50 mg/kg.
Mean determined using one-half the detection limit for non-detect samples.
NA = Not applicable since no PAH was detected.
A significant health concern resulting from exposure to PAHs is their potential
for carcinogenicity, which is chemical-structure dependent. Three-ring PAHs,
including anthracene and fluorene, have not been shown to cause cancer in
laboratory animals [2]. On the other hand, there are seven 4- to 6-ring PAHs
(shown in bold type in Tables 2a and 2b) that are carcinogenic in laboratory
animals [2]. Of the PAHs shown to be carcinogenic, chrysene was found at the
highest concentrations in crude oil (its mean concentration was 28.5 mg/kg oil).
The mean concentrations in crude oil of the other carcinogenic PAHs range from
5.5 mg/kg oil for benz[a]anthracene, to <0.5 mg/kg oil for benzo[k]fluoranthene
and indeno[1,2,3-cd]pyrene. The mean concentration of benzo[a]pyrene was
2.0 mg/kg oil.
The range and mean concentrations of PAHs detected in ten condensate samples
is presented in Table 2b. Overall, PAHs were detected in condensates much less
frequently than in crude oils. Naphthalene, fluorene, and phenanthrene were
detected in >90% of the samples, while dibenz[a,h]anthracene, benzo[a]pyrene,
147
Chapter 10
benzo[k]fluoranthene, indeno[1,2,3-cd]pyrene, and benzo[g,h,i]perylene were

not detected in any sample. Only three of the seven carcinogenic PAHs
(benz[a]anthracene, chrysene, and benzo[b]fluoranthene) were detected in the
condensates. Naphthalene represents >90% of the total priority pollutant PAHs
present in condensates.
Table 3 compares the PAH content of soils from contaminated creosote
production and MGP sites [10,11] with the concentrations of PAHs found in the
ten E&P wastes (sludges/tank bottoms) and six crude oil-containing soils that
were analyzed as part of this study. The concentration of PAHs in E&P wastes
clearly was markedly lower than in site soils affected by creosote or MGP
operations. Furthermore, the distribution of the individual PAHs differ
markedly, with E&P wastes containing very small amounts of 4-, 5-, and 6-ring
PAHs, compared to MGP site soils.
Table 3. Mean PAH concentrations present in creosote, MGP, sludges/tank
bottoms, and E&P soils.
Soil
Concentrations
Creosote
Production
Sites Mean [10]
PAH
(mg/kg)
Naphthalene
1,313
Acenaphthylene
33
Acenaphthene
--Fluorene
650
Anthracene
334
Phenanthrene
1,595
Fluoranthene
682
Pyrene
642
Benz[a]anthracene
--Chrysene
614
Benzo[b]fluoranthene --Benzo[k]fluoranthene --Benzo[a]pyrene
--Indeno[1,2,3,-cd]pyrene--Dibenz[a,h]anthracene --Benzo[g,h,i]perylene ---
Soil
Concentrations
MGP Sites
Mean [11]
(mg/kg)
--2
--225
156
379
2,174
491
317
345
260
238
92
207
2,451
---
Sludges/
Tank
Bottoms
Mean
(mg/kg)
44.00
0.29
6.51
21.09
2.22
55.53
2.31
5.42
2.98
12.16
1.74
0.28
0.97
0.20
0.65
0.73
Soil
Concentrations
E&P Sites
Mean (mg/kg)
0.183
0.017
0.733
0.455
0.214
1.429
0.283
0.869
0.393
1.385
0.199
0.061
0.119
0.024
0.094
0.176
These differences in PAH concentration and distribution are also illustrated in

Figure 3, which compares the mean value for each of the carcinogenic PAHs in
E&P pit sludges/tank bottoms, soils, condensates, and crude oils with those in
coal tar. Coal tar contains approximately 1,000 times more carcinogenic PAHs
than do crude oils, and 10,000 times more than condensates. The most abundant
148
Chapter 10
carcinogenic PAHs in crude oils are benz[a]anthracene and chrysene; while in

condensates, chrysene and benzo[b]fluoranthene are the most abundant.
100000
Mean Values (mg/kg)
10000
1000
100
10
1
Coal Tar
Benz[a]anthracene
Benzo[a]pyrene
Pit sludges/tank
bottoms (10)
Chrysene
Indeno[1,2,3,-cd]pyrene
E&P soils (6)
Crude oils (60)
Dibenz[ a,h]anthracene
Condensates
(10)
Figure 3. Distribution fingerprints of the carcinogenic PAHs found in 60 crude

oils , 10 condensates, 6 oily soils, and 10 pit sludges/tank bottoms from E&P
sites in comparison to those in coal tar [12].
Some differences in the concentrations of 4-6 ring PAHs were noted when
comparing crude oils produced from different regions of the world. For
example, the PAH content is greater, on average, in crude oils produced in
Indonesia and Africa, compared to the oils produced in North America. Further
analysis of this phenomenon revealed that those oils with higher concentrations
of 4-6 ring PAHs are from lacustrine (lakebed) source rocks in Indonesia and
Africa. Lacustrine source crude oils are rare in North America. In addition, the
highest concentrations of 4-6 ring PAHs were present in those lacustrine source
oils that had been biodegraded in the oil reservoir, as shown in Figure 4.
149
Chapter 10
180
160
Biodegraded Lacustrine Crude Oils

>4-ring PAH, mg/kg oil
140
120
Lacustrine Crude Oils

100
80
60
40
20
0
0
10
15
20
25
30
35
40
45
50
API gravity
Figure 4. Concentration of 4-6 ring PAHs versus API gravity of crude oils.
Crude oils from lacustrine sources that have been biodegraded contain the
highest concentrations of >4-ring PAHs.
Figure 4 also illustrates that API gravity is not a good predictor of the
concentration of 4- to 6-ring PAHs that may be present in crude oils. Source
rock and maturity have been reported to correlate with PAH content of crude
oils [13].
Risk-Based Screening Levels for PAHs

Chemical-specific Risk-Based Screening Levels (RBSLs) can be derived for the
individual 16 priority pollutant PAHs. Regulatory agencies in several of the
major oil producing states in the United States have developed RBSLs or
protective concentration levels (PCLs) for PAHs. The approaches and
algorithms used by the agencies to derive Tier 1 levels generally follow those
used by the USEPA in deriving Preliminary Remediation Goals (PRGs) [14] and
Soil Screening guidance levels (SSLs) [15], or those described by the American
Society for Testing and Materials (ASTM) Standard Guide for Risk-Based
Corrective Action Applied at Petroleum Release Sites [16].
Table 4 summarizes the Tier 1 RBSLs developed by state agencies in major oil
producing states for the 16 priority pollutant PAHs. The variability in Tier 1
levels between states is due, in part, to policy decisions regarding acceptable
target risk levels. Different regulatory agencies have adopted different target
risk levels, so that some Tier 1 levels represent those concentrations that would
result in a cancer risk of one-in-a-hundred thousand (1 x 10-5), while
others represent a one-in-a-million (1 x 10-6) target risk. For example, the State
150
Chapter 10
Table 4. Tier 1 RBSLs from major oil producing states showing the concentration of each PAH for soils in residential or
industrial areas, plus those concentrations of crude oil in soil that are protective of groundwater. All units are in mg/kg.
Louisiana2
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Anthracene
Phenanthrene
Fluoranthene
Pyrene
Benz[a]anthracene
Chrysene
Benzo[a]pyrene
Alaska 1
38-43
NA
190-210
240-270
3,900-4,300
NA
NA
1,400-1,500
5.5-6
550-620
17-20
170-200
2.4-3
50-54
5-6
NA
Industrial
44
NA
3,900
3,100
25,000
NA
3,600
2,700
3.6
400
3.6
35
0.36
3.6
0.36
NA
GW
Residential Protect
6.3
1.5
NA
NA
260
220
180
230
1,400
120
NA
NA
200
1,200
150
1,100
0.56
8.6
61
76
0.56
29
5.5
120
0.33
23
0.56
9.2
0.33
540
NA
NA
New Mexico3
Industrial
18,500
NA
27,700
19,600
157,000
14,500
22,100
16,700
21.8
2,150
21.7
21.9
2.19
NA
2.21
NA
GW
Residential Protect
1,790
0.68
NA
NA
2,900
187.95
2,150
196.12
16,900
4,499.81
1,590
270.07
2,340
1,247.59
1,760
1,301.71
9.49
7.48
940
810.27
9.45
25.68
9.52
25.68
0.952
4.74
NA
NA
0.952
3.74
NA
NA
Texas4
Industrial
360
37,000
37,000
25,000
190,000
19,000
25,000
19,000
24
2,400
24
240
2.4
24
2.4
19,000
Residential
220
3,800
3,000
2,300
18,000
1,700
2,300
1,700
5.7
560
5.7
57
0.56
5.7
0.55
1,800
GW
Protect
31
410
240
300
6,900
420
1,900
1,100
18
1,500
60
620
7.6
170
15
46,000
NA = Not available.
1) Cleanup levels based on migration to groundwater assuming over 40-inch and under 40-inch rainfall zones, respectively. Cleanup levels correspond with
Alaska AK101-103AA and AK101-103 methods. Source: ADEC, 2000 [18].
2) Soil screening standards for industrial exposures, non-industrial exposures, and protection of groundwater. Source: LDEQ RECAP, 2000 [19].
3) Industrial and residential direct contact exposures, and Tier 1 levels protective of groundwater assuming a DAF = 1. Source: NMED, 2000 [20].
4) Industrial and residential direct contact exposures and protection of groundwater assuming a 0.5-acre source. Source: TNRCC, 2001 [21].
151
Chapter 10
of Louisiana uses a target risk of 1 x 10-6 for carcinogens and a Hazard Quotient
(HQ) of 0.1 for non-carcinogens, while Texas uses 1 x 10-5 for carcinogens and
a HQ = 1.0. In any case, both of these target cancer risk values lie within the
acceptable target risk range of 1 x 10-6 to 1 x 10-4 set by USEPA for evaluating
contaminated sites under Superfund [17] and are in line with target risk levels
commonly adopted by regulatory agencies for environmental programs in many
states.
Acceptable Total Petroleum Hydrocarbon (TPH)

Levels That Are Protective for PAHs
The maximum concentrations of each of the PAHs detected in crude oils and
condensates listed in Tables 2a and 2b were compared to the lowest state Tier 1
residential RBSL listed in Table 4. This comparison was done to determine the
acceptable oil concentration in soil [in terms of its TPH content] that would
result in no PAH being above any single residential or groundwater protection
Tier 1 level.
In reality, for all non-carcinogenic PAHs (except
benzo[g,h,i]perylene), the lowest Tier 1 RBSL is for groundwater protection.
Tier 1 RBSLs that are protective of direct residential exposure to impacted soils
are higher (i.e., less stringent) than groundwater protection levels. The lowest
Tier 1 RBSLs for carcinogenic PAHs and benzo[g,h,i]perylene are based on
direct residential exposure to impacted soils.
The following formula was used to estimate acceptable TPH levels in soil that
would assure that all PAH levels were below their respective individual RBSLs:
Acceptable TPH (% Oil in Soil) = RBSL (mg PAH/kg Soil)/Coil (mg PAH/kg Oil) * 100
The acceptable TPH level was estimated for each oil or condensate, and the
results are shown in Table 5. For the crude oils, the estimated acceptable TPH
levels ranged from 1.4 to >100%, except for the case of naphthalene. For the
condensates, the estimated acceptable TPH levels in soil range from 28 to
>100%. Again, naphthalene is an exception, resulting in an acceptable
concentration of condensate in soil as low as 0.012%. This TPH level is
protective of groundwater resources. Acceptable TPH levels for direct
residential contact with impacted s oils would be higher.
Naphthalene may be present in some crude oils and condensates at
concentrations that exceed Tier 1 PAH levels even at low TPH levels.
Naphthalene may be of particular concern for protecting groundwater resources.
However, the acceptable oil and condensate concentrations provided in Table 5
are for fresh spills. Natural attenuation of naphthalene due to weathering
(volatilization and biodegradation) may occur quickly at some spill sites. Also,
Tier 1 RBSLs for naphthalene do not consider the impact of Raoults Law as
described in Chapter 9. For these reasons, the preferred method for assessing
152
Chapter 10
risk due to naphthalene is to analyze specifically for naphthalene in soil, rather

than setting very low Tier 1 levels for TPH. This is consistent with the approach
for benzene described in Chapter 11.
Fifteen of the priority pollutant PAHs should not be present above any
residential Tier 1 level at either fresh crude oil or condensate spill sites, as long
as TPH levels are less than 1.0% (10,000 mg/kg). Therefore, there is no need to
routinely use USEPA Method 8270 to quantify these 15 PAHs at E&P spill
sites. The only PAH that may be present at concentrations of concern is
naphthalene, and it can be analyzed using USEPA Method 8260 for volatiles.
Eliminating the use of USEPA Method 8270 for all priority pollutant PAHs at
E&P spill sites is cost-effective, while still being protective of human health.
Table 5. Comparison of lowest Tier 1 levels in major oil producing states with
the maximum concentration of PAH observed in crude oils and condensates.
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Anthracene
Phenanthrene
Fluoranthene
Pyrene
Benz[a]anthracene
Chrysene
Benzo[a]pyrene
Maximum Observed
Lowest Tier 1 Concentration
(mg/kg Oil)
RBSL From
Crude
Table 4
(mg/kg Soil) Oils
Condensates
0.68
3,700
5,700
410
ND
9.2
187.95
58
12
180
380
82
120
17
27
270.07
916
250
200
26
11
150
82
12
0.56
38
0.78
61
120
5.5
0.56
14
2
5.5
7
ND
0.33
7.7
ND
0.56
1.7
ND
0.33
9.2
ND
1800
9.6
ND
Acceptable
Concentration for a
Fresh Spill (% Oil or
Condensate in Soil)
Crude Oil Condensate
in Soil
in Soil
0.018
0.012
>100
>100
>100
>100
47.3
>100
>100
>100
29.5
>100
>100
>100
>100
>100
1.4
71.8
50.8
>100
4.0
28.0
78.5
>100
4.2
>100
32.9
>100
3.5
>100
>100
>100
ND = Below detection limits.
This examination of PAH content of oils and condensates applies to situations

involving a single oil spill incident or single application of oil. Sites that have
received multiple applications of crude oil or oil-contaminated soils, e.g., a land
farm site or an old pit, would require additional consideration.
Summary
Sixty crude oils and ten condensates were analyzed for their concentrations of
16 priority pollutant PAHs. Screening of the human health risk associated with
153
Chapter 10
soil containing crude oil or condensate showed the risk for 15 of the 16 PAHs
was not significant at TPH concentrations well above 10,000 ppm, the current
TPH soil management level in many states. Naphthalene is the only PAH that
may be present in crude oils or condensates at concentrations that may pose a
risk at oil concentrations of 1% or lower. The limiting exposure pathway for
naphthalene is leaching from soil to groundwater and protection of groundwater
resources, rather than direct residential exposure to contaminated surface soils.
Overall, this work indicates that the low levels of PAHs in crude oils are
unlikely to be a major risk management consideration at crude oil or condensate
spill sites. This constitutes compelling evidence that routine analyses for all
PAHs at E&P crude oil or condensate spill sites are unnecessary. However,
naphthalene should be analyzed as part of an USEPA 8260 analysis along with
the volatiles benzene, toluene, ethyl benzene, and xylenes.
References
1.
Blumer, M., Polycyclic Aromatic Compounds in Nature, Scientific

American, 234 (3) (1976).
2.
Agency for Toxic Substances and Disease Registry, Toxicological Profile

for Polyaromatic Hydrocarbons, U.S. Department of Health & Human
Services (1995).
3.
Bence, A.E., et al., Fingerprinting Hydrocarbons in the Biological

Resources of the Exxon Valdez Spill Area, American Society for Testing
and Materials (1995).
4.
Lide, D.R., CRC Handbook of Chemistry and Physics , 77th edition, CRC
Press, Inc., Boca Raton (1996-1997).
5.
MacKay D., Shiu, W.Y., and Ma, K.C., Illustrated Handbook of PhysicalChemical Properties and Environmental Fate for Organic Chemicals,
Vol. 2, Lewis Publishers, Chelsea, MI (1992).
6.
Karcher, W., Spectral Analysis of Polycyclic Aromatic Compounds, Vol. 2,

Kulwer Academic Publishers, The Netherlands (1988).
7.
API Publication Number DR53, Characterization of Exploration and

Production Associated Wastes (1996).
8.
U.S. Environmental Protection Agency, Draft Associated Waste Report:

Crude Oil Tank Bottoms and Oily Debris (1994).
9.
ASTM Publication Number PS-104-98, Standard Provisional Guide for

Risk-Based Corrective Action (1998).
10. Ellis, B., et al., Bioremediation of a Creosote Contaminated Site,

Environmental Technology, 12, 447-59 (1991).
154
Chapter 10
11. Bewley, R., et al., Microbial Cleanup of Contaminated Soil, Chemistry

and Industry, 23,778-83 (1989).
12. Nishioka, M., et al., Structural Characteristics of Polycyclic Aromatic
Hydrocarbon Isomers in Coal Tars and Combustion Products,
Environmental Science and Technology, 20, 1023-1027 (1986).
13. Requejo, A.G., Sassen, R., McDonald, T., Denoux, G., Kennicutt, M.C.,
and Brooks, J.M., Polynuclear Aromatic Hydrocarbons (PAH) as
Indicators of the Source and Maturity of Marine Crude Oils, Organic
Geochemistry, 24, 1017-1033 (1996).
14. U.S. Environmental Protection Agency (USEPA), Region IX Preliminary
Remediation Goals (PRGs), November 2000.
15. EPA Office of Emergency and Remedial Response, Soil Screening
Guidance: Technical Background Document, USEPA/540/r-95/128,
Washington, D.C. (1996).
16. American Society for Testing and Materials, Standard Guide for
Corrective Action Applied at Petroleum Release Sites, ASTM E 1739-95,
West Conshohocken, Pennsylvania (1995).
17. U.S. Environmental Protection Agency (USEPA), Risk Assessment
Guidance for Superfund. Vol. 1. Human Health Evaluation Manual (Part
A), USEPA/540/1-89/002, Environmental Protection Agency Region 1,
Washington, D.C. (1989).
18. Alaska Department of Environmental Conservation, Handbook for
Conducting Cleanups of Contaminated Sites and Regulated Underground
Storage Tanks Under the Voluntary Cleanup Program, March 2000.
19. Louisiana Department of Environmental Quality (LDEQ), Risk
Evaluation/ Corrective Action Program, June 2000.
20. New Mexico Environment Department (NMED), New Mexico
Underground Storage Tank Bureau Guidelines for Corrective Action,
March 2000.
21. Texas Natural Resource Conservation Commission (TNRCC), Texas Risk
Reduction Program Rule. Tier 1 Protective Concentration Limit Tables,
March 2001.
155
Chapter 11
An Evaluation of Benzene Risk

William G. Rixey
University of Houston
Introduction
Benzene is commonly found in the environment from both human activities and
natural processes. Benzene was first discovered in 1825 and isolated from coal
tar in 1849, while today it is manufactured mostly from petroleum sources.
Benzene is used by industry to make other chemicals such as ethylbenzene for
plastics manufacture, cumene for resins, and cyclohexane for nylon and
synthetic fibers [1]. Natural sources of benzene in the environment include
volcanoes, forest fires, and crude oil seeps. Benzene occurs naturally in most
crude oils, is a byproduct of oil refining processes, and also occurs in natural gas
production condensates.
Benzene is a known human carcinogen. In workers, long-term exposure to high
concentrations of benzene in air has been shown to cause cancer of the bloodforming organs. In laboratory animals, benzene has been shown to produce
several types of cancer following oral or inhalation exposure. There are still
questions concerning both the mechanisms of benzene carcinogenesis and the
most appropriate models for developing human risk estimates. These issues are
actively being studied and debated in the scientific community. In addition to
cancer, benzene is also known to produce other adverse health effects, again
principally on the blood-forming organs, although neurological and reproductive
effects may also be of concern [1]. Most people are exposed to a small amount
of benzene every day, mainly via inhalation of vapors from commercial products
such as glues, paints, cigarette smoke, and vehicle exhaust. People may come
into contact with benzene through the inhalation, ingestion, or dermal contact
exposure pathways.
Most upstream regulatory programs (such as those in the States of Texas and
Louisiana) do not routinely require benzene analysis of exploration and
production (E&P) site soils and do not routinely set regulatory limits for
benzene in soil. Upstream regulatory agencies in California, New Mexico, and
Michigan are exceptions and do require benzene analyses for soils at E&P sites.
Regulatory limits for benzene in soil are routinely set in downstream regulatory
programs , such as those with jurisdiction over underground storage tank (UST)
sites . Most often, these are based on Tier 1 Risk Based Screening Levels
(RBSLs) developed for protecting groundwater resources. In developing RBSLs,
a number of fate and transport assumptions are typically used that are now
known to be overly conservative for benzene. For example, Tier 1 RBSLs have
156
Chapter 11
historically assumed that benzene in a complex mixture of petroleum

hydrocarbons in soil behaves in the same way it would if it were the only
chemical present in soil, and that there are no losses of benzene due to
volatilization or biodegradation over time.
This study was conducted to improve the fate and transport assumptions
typically used to derive RBSLs for benzene in soil. RBSLs are developed that
take into consideration the attenuation of benzene in the vadose zone, as well as
the presence of the complex petroleum mixture (expressed in terms of TPH) in
soil. Additional attenuation of benzene in groundwater is not considered. The
potential risk that benzene might pose at E&P sites is then evaluated by
comparing these RBSLs to two estimates of potential benzene levels in E&P site
soils. The first estimate is based on benzene levels found in several
unweathered crude oils and condensates . The second is based on limited field
data for actual benzene levels measured in E&P site soils following typical
emergency response activities after spill events .
Benzene Concentrations in Crude Oils and Condensates

Sixty-nine unweathered crude oils and fourteen unweathered condensate
samples were analyzed for volatile aromatic hydrocarbons, including benzene,
toluene, ethylbenzene, and xylenes (BTEX) using purge and trap gas
chromatography coupled with mass spectrometry (GC/MS). Samples were
analyzed following a procedure based on the United States Environmental
Protection Agency (USEPA) Method 8260A [2].
Figure 1 illustrates the sample locations for the 69 crude oils and 14
condensates. The American Petroleum Institute (API) gravity range for the
crude oils in this study is 9 to 46o, and the range is 45 to 70 for the condensates.
While all of the samples were analyzed for BTEX as discussed above, only the
benzene values are presented here . Toluene, ethylbenzene, and xylenes (TEX)
do not typically present a risk management concern at petroleum release sites.
They are non-carcinogenic compounds and they are addressed as part of the
petroleum mixture as a whole. TEX are included in the non-carcinogenic TPH
RBSLs presented in Chapter 8. The analytical results for TEX are provided in
Chapter 4.
157
Chapter 11
Figure 1. Sample locations for the crude oils and condensates. Twelve of the
condensates were from the United States (excluding Alaska and Hawaii).
As shown in Table 1, the highest observed concentration of benzene in the 69
crude oils was 5900 mg/kg of oil or 0.59 wt %, and the mean concentration of
benzene in the crude oils was 1,340 mg/kg. Two crude oils contained less than
1.2 mg/kg benzene (the detection limit for the analytical method). In general,
higher API gravity crude oils and condensates tend to contain more benzene as
shown in Figure 2. The condensates contained more benzene than the crude
oils, with the maximum concentration being 35,600 mg/kg of condensate
(3.56 wt %). The mean concentration of benzene for the 14 condensates was
10,300 mg/kg. There is roughly 10 times more benzene on average in the
analyzed condensates than in the analyzed crude oils.
Table 1. Concentrations of benzene in crude oils and condensates analyzed in
this study.
Concentration of Benzene (mg/kg Oil)
# of Samples
69 Crude Oils
14 Condensates
API Gravity
Range ()
8.846.4
4570.1
Mean
1,340
10,300
Median
780
6400
Minimum
ND*
1470
Maximum
5900
35,600
Number of
Samples With
Benzene = ND
2
0
ND = Non-detect, with the sample detection limit = 0.32 mg benzene/kg oil.
158
Benzene concentration in oil (mg/kg)
Chapter 11
105
104
103
102
101
Benzene Conc. in Oil Data
Regression line
100
10-1
0
10
20
30
40
50
60
70
API Gravity
Figure 2. Benzene concentrations versus API gravity for 61 crude oils and 14
condensates (API gravity data were unavailable for 8 crude oils).
Benzene RBSLs for Groundwater Protection

Groundwater protection RBSLs for benzene in soil were developed for the
scenario shown in Figure 3. In this scenario, a surface impoundment or a soil is
impacted from a surface spill of oil (or condensate) in which the oil is confined
to the unsaturated zone and does not reach groundwater, i.e., there is no freephase oil at the water table. It is assumed that some response to the spill has
already occurred and that the extent of contamination has been delineated such
that the depth of contamination and level of contamination are known. The
source of benzene contamination is confined to a layer of thickness, d, and the
bottom of the contaminated layer is a distance, H, from the groundwater table.
159
Chapter 11
Figure 3. A conceptual site model showing a crude oil or condensate

contaminated soil in the unsaturated zone and the various dilution and
attenuation processes that occur during transport to groundwater.
In general, dilution and attenuation of benzene, or other chemicals , emanating
from a source can occur in the unsaturated zone, a groundwater mixing zone,
and in the groundwater downstream of the source. A typical RBSL calculation
assumes an infinite source of the chemical of concern, which means there are no
losses over time due to volatilization, leaching, or biodegradation in the
hydrocarbon impacted layer. In addition, biodegradation in the zone between
the bottom of the impacted soil and the groundwater table is not typically
considered. These assumptions are overly conservative for benzene, because
benzene is volatile and readily biodegradable, provided that oxygen does not
limit the rate of biodegradation. Accordingly, the approach presented here for
developing groundwater protection RBSLs for benzene in soil considers
attenuation effects in the unsaturated zone, including a conservative accounting
of degradation in the unsaturated zone between the bottom of the contaminated
zone and the water table, as well as losses in the source of contamination due to
volatilization and leaching. These are accounted for in an unsaturated zone
dilution attenuation factor (DAF), DAFunsat.
In addition to these attenuation considerations, the enhanced soil-water
partitioning that occurs for benzene, due to the presence of a complex petroleum
mixture in soil is also considered (i.e., the petroleum mixture keeps more of the
benzene in the oily soil phase). Present approaches for calculating individual
chemical RBSLs assume that partitioning occurs to native soil organic matter
only.
160
Chapter 11
The RBSL for benzene in soil that is protective of groundwater, RBSLs-gw, was
calculated using the following equation:
RBSL s gw = CT = K T * DAF o * C R
(1)
where:
CT = The total concentration of benzene in soil based on the groundwater
exposure pathway (g/g-soil)
CR = The acceptable concentration of benzene in groundwater at the
groundwater receptor (g/cm3-water)
CR is either a risk-based screening level for water, RBSLw, or a groundwater
regulatory standard (in the United States, the maximum contaminant level for
benzene in groundwater is 5 x 10-9 g/cm3-water; the State of New Mexico has a
groundwater standard = 1.0 x 10-8 g/cm3-water).
If appropriate, RBSLw (g/cm3-w) can be calculated for benzene, a carcinogen,
according to the following equation:
RBSL w =
TR * BW * AT c * 365 * 10 6
IR w * ED * EF * SFo
(2)
where:
Target excess individual lifetime cancer risk (10-5)
Adult body weight (70 kg)
Averaging time for carcinogens (70 years)
Oral cancer slope factor [mg/kg-day]-1 (0.029 for benzene)
Adult daily water ingestion rate (2 L/day)
Exposure frequency for residents (350 days/year)
Exposure duration for adult residents (30 years)
The total partition coefficient for benzene. It is the ratio of total soil
concentration to pore water concentration in the source zone of the
contamination (cm3-water/g-soil).
DAFo = The overall dilution attenuation factor (unitless) which is defined as:
TR
BW
ATc
SFo
IRw
EF
ED
KT
=
=
=
=
=
=
=
=
DAFo = DAFunsat * DAFmix * DAFsat
(3)
where:
DAFunsat = Cs/Cunsat = Unsaturated zone dilution attenuation factor (unitless)
161
Chapter 11
DAFmix = Cunsat/Cmix = Groundwater mixing zone dilution attenuation factor

(unitless)
DAFsat = Cmix/CR = Dilution attenuation factor in groundwater downstream of
the source (unitless)
Cs
= Concentration in pore water at the source of contamination (g/cm 3 - w)
Cunsat
= Concentration in pore water at the bottom of the unsaturated zone
(g/cm3-w)
Cmix
= Concentration in groundwater at the downstream edge of the mixing
zone (g/cm3-w)
Equations for determining the overall soil-water partition coefficient and the
various DAFs are presented in the paragraphs below.
Soil-Water Partition Coefficient, KT
The overall soil-water partition coefficient, KT, is given by:
KT =
CT w + K o b + K a a + K o o
=
CS
b
(4)
where:
CT
CS
b
w
Kd
Ka
= Total concentration of chemical in soil (g i/g soil)

= Concentration in pore water at the source of contamination (g i/cm3-w)
= Soil bulk density (g -soil/cm3-soil)
= Volumetric water content in vadose zone soils (cm3-w/cm3-soil)
= Soil-water sorption coefficient for chemical (cm3-w/g-soil)
= Air-water partition coefficient (dimensionless Henrys Law constant) for
chemical (cm3-w/cm3-air)
a = Volumetric air content in vadose zone soils (cm3-air/cm3-soil)
Ko = Oil-water partition coefficient (cm3-w/cm3-oil)
o = Volumetric oil content in vadose zone soils (cm3-air/cm3-soil)
Note that in Equation (4) partitioning of the chemical to residually trapped oil in
the soil is included. This represents additional partitioning that occurs when a
residual oil phase is present. Thus the levels of benzene that are acceptable in
the soil depend on the oil content in the soil. The oil-water partition coefficient
can be determined from Raoults Law [3,4,5] as:
Ko =
o MWi
S i MWo
(5)
where:
162
Chapter 11
o
MW i
MW o
Si
=
=
=
=
Density of oil phase (g-o/cm3-o)

Molecular weight of contaminant (g i/mole)
Molecular weight of oil phase (g -o/mole)
Solubility of pure chemical in water (g i/cm3-w)
The volumetric oil content in the soil can be related to the residual TPH
concentration with the following equation:
C
o = b TPH
o
(6)
where:
CTPH = Residually trapped TPH concentration in soil (g/g-soil)
Substituting Equations (5) and (6) for Koo, Equation (4) becomes:
KT =
w + K d b + K aa +
b
b MWi CTPH
S i MWo
(7)
Thus, the overall partition coefficient, KT, will be dependent upon the level of
residual TPH in the soil. (Note that the residual TPH level is the amount in
excess of the sorbed TPH level which is nominally <100 mg/kg for a low
organic carbon content soil.)
Summary of DAFunsat
The attenuation in the unsaturated zone is due to the following factors:
1) Biodegradation of the contaminant in the region beneath the source of
contamination and the groundwater table.
2) Depletion of the concentrations in the source of contamination due to
losses associated with leaching, volatilization, and biodegradation.
A commonly used approach for modeling vadose zone transport is that of nl
et al. [6] which uses the equation of van Genuchten and Alves [7]. This
equation is also the basis for modeling unsaturated zone transport in the
computer model VADSAT [6]. To determine DAFunsat values for the RBSLs
presented here a simpler model was used. This model is based on treating the
contaminated zone and the unsaturated zone beneath the contaminated zone (see
Figure 3) as two separate, completely mixed zones. (The nl model treats the
163
Chapter 11
source zone as a completely mixed zone, but models advection and dispersion in
the region below the source zone.) This completely -mixed model approach was
used here because it is computationally easy to use (it is possible to develop a
simple algebraic expression for DAFunsat) and retains the essential parameters of
the nl et al. model. For completely -mixed conditions, the DAFunsat can be
determined from the following equation:
DAFunsat =
H b K T ,unsat
u
(8)
where:
u
+ unsat
H
=
b K T , unsat
(9)
and
u
+ S
d
b K T ,S
V +
(10)
DAFunsat in Equation 8 is equal to the ratio Cso/Cunsat, max where:

Cso
Cunsat, max
= The initial concentration in the pore water at the source of

contamination (g i/cm3-w)
= Maximum pore water concentration at the bottom of the
unsaturated zone (g i/cm3-w)
The parameter (day -1) represents the effect of biodegradation in the

unsaturated zone on DAFunsat and the parameter (day-1) represents the effect of
the various source losses on DAFunsat. The following parameters that make up
and are:
u
H
d
unsat
S
= The infiltration rate (cm/day)

= Distance from the bottom of the contaminated source region to the
water table (cm)
= Depth of the source of contamination (cm)
= Degradation constant in unsaturated zone beneath the source zone
(day -1)
= Degradation constant in the source zone (day -1)
164
Chapter 11
KT,unsat = The overall soil-water partition coefficient for the unsaturated zone
(cm3-water/g-soil)
KT,S
= The overall soil-water partition coefficient for the source zone (cm3water/g-soil)
V
= Volatilization rate constant for the source zone (day -1)
b
= Bulk density of the soil (g-soil/cm3-soil)
The volatilization rate constant, V, can be estimated from:
V =
2 DwT , S
4d 2
(11)
where:
DwT,S = Total effective diffusion coefficient defined in terms of a chemicals
concentration gradient in water (cm2/day)
DwT,S is defined mathematically as:
DwT , S = Dw, eff + K a Da , eff + K o Do, eff
(12)
The effective diffusion coefficients were determined as follows [8]:

Dw, eff = D wm
w3.33
a 3.33
o 3 .33
;
D
=
D
;
and
D
=
D
a
,
eff
am
o
,
eff
om
2
2
2
(13)
where Dwm , Dam , and Dom are the molecular diffusion coefficients (cm2/sec) of
the given chemical in the water, air, and residual oil phases, respectively. The
following values were used for benzene: Dwm = 1.10E-05 cm2/sec [9], Dam =
0.093 cm2/sec [9], Dom = 3.4E-05 cm2/sec [10]. In general, the contribution of
the oil phase diffusion term to DwT,S was not significant.
The expression for the first order volatilization rate constant, V, was determined
from the solution for transient diffusion at long times from a slab of thickness, d,
with a pore water concentration equal to zero at the top surface and zero flux at
the bottom surface [11]. We have therefore assumed that there is no additional
resistance to mass transfer at the soil-air interface. The boundary layer mass
transfer resistance will be small relative to the diffusion resistance in the soil, so
it is reasonable to neglect this resistance. If an overburden layer exists, then its
thickness can be added to the parameter d in Equation 11. This assumes that the
overburden properties are similar to the source region. This would be a
conservative assumption with respect to source depletion, since the partitioning
165
Chapter 11
would be lower in the overburden layer than in the source region due to the
presence of residual hydrocarbon in the source versus the overburden layer.
Biodegradation of benzene in subsurface soils can be limited by the mass
transfer of oxygen. As a result, the first step in evaluating attenuation of
benzene due to degradation was to determine whether there would be sufficient
oxygen present from the bottom of the contaminated source region downward
through the unsaturated zone to the groundwater table. When oxygen
concentrations are above a threshold concentration, then a conservative first
order reaction rate constant can be used to estimate the attenuation of benzene in
the unsaturated zone. In developing the benzene RBSLs, a conservative first
order pore water-based biodegradation rate constant = 0.01 day-1 was assumed
when the source thickness was less than a critical value.1 For a sandy soil this
critical source thickness was estimated to be d 5 ft (calculations are not
presented here). A pore water-based biodegradation rate constant = 0.01 day-1
translates to a soil based degradation constant, unsat = 0.00094 day-1 for a sandy
soil with a volumetric water content, w = 0.094 cm3-w/cm3-soil.
Note that Equation (8) assumes that the DAFunsat is based on the maximum
aqueous concentration that will reach the groundwater table. It is therefore
conservative, since the average concentration in water observed over a typical
exposure period would be significantly lower.
Summary of DAFmix
The DAF in the mixing zone can be determined from the following equation:
DAF mix = 1 +
U gw gw
IL
(14)
where:
Ugw = Groundwater Darcy velocity (cm/year)
I = Infiltration rate of water through soil (cm/year)
L = Length of source area parallel to groundwater flow direction (cm)
A review of aerobic degradation studies of benzene inferred 1st order

degradation rate constants ranging from 0.02-2 day-1 for high benzene
concentrations, i.e., 100 mg/L [12]. These rate constants were based on water
phase concentrations and were obtained from a review of several laboratory
microcosm and column studies and some field measurements for which
hydrocarbon and oxygen concentration profiles in the unsaturated zone were
determined. Apparent 1st order degradation constants will be higher than these
values for lower benzene pore water concentrations [12].
166
Chapter 11
gw = Groundwater mixing zone thickness (cm)

A DAFmix value = 8.7 was used to calculate benzene RBSLs. This is the value
used in the recently developed New Mexico UST Guidelines for Corrective
Action [9]. For comparison, using the ASTM [13,14] default parameters for
Ugw, I, L, and gw yields a DAFmix = 12.1.
Summary of DAFsat
A DAFsat value = 1 has been assumed which is equivalent to no attenuation
downstream of the source due to dispersion and degradation. This would be
valid if the receptor is located at the downstream edge of the source. If the
receptor is located some distance away from the source, DAFsat will increase
because of dispersion and biodegradation of the chemical in groundwater.
Values of DAFsat developed for a recently developed soil screening guidance
program are shown in Table 2. Biodegradation in groundwater was neglected
for the values shown in Table 2. Accounting for the typical biodegradation rates
for benzene that occur in groundwater will result in higher DAFsat values than
those shown in Table 2. Equations such as that of Domenico [15] with
appropriate soil and chemical parameters can be used for calculating screening
level estimates of DAFsat.
Table 2. Dilution Attenuation Factors (DAFsat) (1).
Distance From Edge of
Mixing Zone (Ft)
0
50
100
150
250
500
1000
DAFsat
(Without Degradation)
1.0
1.1
2.2
4.2
11
41
163
1) From Table 4-14, New Mexico Environmental Department Guidelines for

Corrective Action [9].
Benzene RBSLs-gw Curves
Benzene RBSLs-gw were calculated using the equations described above for
several values of TPH, H (distance from the bottom of the contaminated source
zone to the water table) and a constant value of d (thickness of the contaminated
source zone). No degradation in the groundwater (saturated zone) downstream
of the source area was included in these calculations. If degradation were
167
Chapter 11
included (meaning DAFsat > 1), the RBSLs would be greater than those shown in
Figure 4.
Benzene in soil (mg/kg)
100
Unweathered cond.
(median Ben conc.)
H=100 ft
H=50 ft
10
Unweathered crude
(median Ben conc.)
H=15 ft
H=5 ft
E&P Field Data

(after emergency
response activities)
Benzene RBSL
(d=5 ft.)
0.1
H=0
0.01
0.001
10
100
1000
10000
TPH in soil (mg/kg)
Figure 4. Comparison of soil RBSLs for benzene with a) predicted soil levels
for unweathered crude oils and condensates and b) actual soil levels after
emergency response activities at a few E&P sites. Open (white) symbols
represent detection levels for benzene. Soil RBSLs for benzene are for the
groundwater exposure pathway. Calculations were based on a sandy soil type
and include vadose zone attenuation due to volatilization and leaching from the
source and biodegradation beneath the source (see Figure 3). H = the distance
from the bottom of the contaminated soil zone to the groundwater table.
The calculation of RBSLs considers the following important factors: 1)
degradation in the unsaturated zone between the source of contamination and the
groundwater table, 2) source losses due to volatilization and leaching, and 3)
enhanced partitioning of benzene in soils due to the presence of TPH.
Parameters that determine the relative importance of these factors include the
depth to groundwater, the thickness of the source of contamination, and the soil
type. All of these three factors significantly contribute to higher acceptable
levels of benzene in soil than would be acceptable if attenuation and increased
partitioning in the vadose zone were not considered.
The RBSL curves presented in Figure 4 are based on a sandy soil type; a
biodegradation constant, unsat = 9.4E-04 day-1; a hydrocarbon impacted layer of
thickness, d=5 ft; and a groundwater standard = 0.010 mg/L. Values of other
parameters for determining the curves in Figure 4 are listed in Table 3.
168
Chapter 11
Table 3. Parameters used for benzene DAFunsat and RBSLs-gw calculations.

Soil or Contaminant
Parameter
u (cm/yr)
(cm3-pores/cm3-soil)
w (cm3-w/cm3-soil)
o (cm3-o/cm3-soil)
a (cm3-a/cm3-soil)
Dwm (cm2/sec)
Dam (cm2/sec)
Dom (cm2/sec)
Ka (cm3-w/cm3-a)
Ko (cm3-w/cm3-o)
foc (g oc/g soil)
Koc (cm3-w/g oc)
b (g soil/cm3 soil)
o (g oil/cm3 oil)
DwT,S (cm2/sec)
KT,S (cm3-w/g soil)
KT,unsat (cm3 -w/g soil)
MW i
MW o
S (day-1)
unsat (day-1)
d (ft)
Source Layer (1)

20
0.349(2)
0.094(3)
0.019(1)
0.236(1)
1.10E-05
0.093
3.40E-05
2.20E-01
2.01E+02
2.00E-03
83
1.73(2)
0.90
1.36E-03(1)
2.47(1)
78
200
0
5
Unsaturated Zone
Layer Beneath Source
20
0.349(2)
0.094(3)
0.0
0.255
1.10E-05
0.093
2.20E-01
2.00E-03
83
1.73(2)
0.252
78
9.4E-04
-
1) Values shown are for TPH = 10,000 mg/kg-soil.

2) From Brakensiek et al. [16].
3) Determined from u = 20 cm/yr and Brooks -Cory parameters from
Brakensiek et al. [16] and saturated hydraulic conductivities from Carsel
and Parrish [17].
In Figure 4 RBSLs are shown for benzene as a function of TPH in the soil and
for distance to groundwater, H. The following key points can be made
regarding this figure:
Effect of TPH on Benzene RBSL s-gw
The presence of TPH increases the benzene RBSL due to increased partitioning.
At short depths to groundwater, the presence of TPH has the greatest effect on
the RBSL. When only the sorption of benzene to soil particles is considered
(TPH <100 mg/kg), for H = 0 the RBSL = 0.022 mg/kg. A level of TPH of
169
Chapter 11
10,000 mg/kg results in an increase in the RBSL to 0.21 mg/kg. There is less
effect of TPH on RBSLs at greater depths, H, because higher TPH levels result
in relatively less attenuation from source losses.
Effect of Depth to Groundwater, H, on Benzene RBSLs-gw
When H = 0, there is no attenuation in the unsaturated zone (DAFunsat = 1). As H
increases, attenuation occurs between the bottom of the contaminated zone and
the groundwater table. This is a result of degradation over depth H and of losses
due to volatilization and leaching in the hydrocarbon impacted layer of
thickness, d. As H increases, attenuation increases, and the RBSL increases.
For d=5 ft, the increase in RBSL is due to increased degradation with
contributions from losses due to volatilization and leaching in the contaminated
region.
Benzene RBSLs for Surface Soil Exposure

Benzene RBSLs for exposure of commercial workers to surficial soils via the
pathways of ingestion, inhalation of vapors and particulates, and dermal contact
(RBSLss) were also determined for comparison with the groundwater based
RBSLs-gw values and are shown in Table 4. Methods for determining RBSLs s
were consistent with ASTM E1739-95 and E2081-00 guidelines [13,14].
Table 4. Benzene risk-based soil screening levels for exposure of a commercial
worker to surficial soil (1).
Thickness of Impacted
Layer, d (Ft)
2
5
10
RBSLss
(mg/kg Soil)
484
290
174
1) Surficial soil pathways include: ingestion, inhalation (vapor emission and

particulates), and dermal contact.
The RBSLss values in Table 4 increase as the thickness of the impacted layer (d)
decreases, because a smaller thickness (d) results in a lower exposure to
benzene. These RBSLss values are greater than the benzene soil RBSL values
shown in Figure 4 based on the groundwater exposure pathway (RBSLs-gw).
This indicates that groundwater protection is likely to be the major risk
management concern for benzene at most sites.
170
Chapter 11
Comparison of RBSLs With

Expected Benzene Levels at E&P Sites
Figure 4 also presents lines for predicted benzene soil levels that would
correspond to varying levels of E&P site contamination by unweathered
condensates and crude oils. The lines are based on the median concentrations of
benzene observed for the 69 crude oils and the 14 condensates for which
composition data were summarized in Table 1. Figure 4 also presents field data
from various E&P sites for benzene soil concentrations, plotted versus the
corresponding TPH soil levels for these sites. Most of these data are from sites
impacted by crude oils, but data from a few condensate impacted sites are also
included. For these sites, soil samples were obtained soon after emergency
response activities were completed.
Comparisons of the E&P field data with the unweathered crude oil and
condensate data indicate that typical emergency response activities significantly
reduce the levels of benzene in soils. Figure 4 also indicates that sites impacted
by condensates and crude oils may not exceed benzene groundwater standards
even for short distances to the water table (H). However, benzene soil levels
that correspond to unweathered crude oils and especially to unweathered
condensates, may present a risk to groundwater at certain TPH levels and depths
to groundwater. It should be noted that the benzene RBSLs presented in this
chapter do not account for attenuation in the saturated zone, which can be
significant when the point of compliance is downstream of the source.
Summary
An improved approach to developing RBSLs for benzene in soil that are
protective of groundwater was developed that makes use of more realistic fate
and transport assumptions than are typically used in most Tier 1 calculations.
Attenuation effects were considered, including a conservative accounting of
minimal biodegradation in the unsaturated zone between the bottom of the
contaminated zone and the water table, as well as losses in the source of
contamination due to volatilization and leaching. In addition to these
attenuation considerations, the enhanced soil-water partitioning that occurs for
benzene due to the presence of TPH in soil was also considered. All of these
factors significantly contribute to higher acceptable levels of benzene in soil
than would be estimated if attenuation and increased partitioning in the vadose
zone were not considered. Additional attenuation in the saturated zone was not
considered but would further increase acceptable levels of benzene in some
cases.
Benzene RBSLs were found to depend on some key parameters: 1) depth to
groundwater (H), 2) thickness of oil impacted layer (d), and 3) level of TPH in
the soil. These parameters determine the amount of attenuation of benzene in
171
Chapter 11
the vadose zone and the decreased partitioning of benzene to soil when TPH is
present. An easy to use figure was presented that can be used to select the
appropriate benzene RBSL for an individual site if the above factors are known.
This figure illustrates that RBSLs increase by a factor of 10 to 1000 when
vadose zone attenuation and increased partitioning (water to soil) due to TPH
are taken into account.
The potential risk posed by benzene at E&P sites was also evaluated. Overall,
the major risk management concern for benzene at most E&P sites is likely to be
due to its potential to impact groundwater and not due to direct commercial
worker exposure to impacted surface soils. Benzene may present a risk to
groundwater at some E&P sites. The potential risk will depend on the type of
oil (crude oil or condensate) spilled, the depth to groundwater, the thickness of
contamination, the level of TPH in the soil, and the extent of weathering of
benzene from soils that results from any emergency response activities.
The benzene RBSLs presented in this chapter are illustrative of screening levels
that could be used in conjunction with TPH RBSLs to decide if further
corrective action is required at a given site. To use screening levels such as
these for benzene it is assumed that some response to a spill has already
occurred and that the extent of contamination has been delineated such that the
depth of contamination and level of contamination of benzene and TPH are
known.
References
1.
Agency for Toxic Substances and Disease Registry, Toxicological Profile

for Benzene, U.S. Department of Health and Human Services, Public
Health Service (1997).
2.
United States Environmental Protection Agency (USEPA), Test Methods

for Evaluating Solid Waste (SW -846), Vol. 1B, Method 8260, U.S.
Environmental Protection Agency, Office of Solid Waste and Emergency
Response, Washington, D.C. (1986).
3.
Cline, P.V., Delfino, J.J., and Rao, P.S.C., Partitioning of Aromatic

Constituents Into Water From Gasoline and Other Complex Solvent
Mixtures, Environ. Sci. Technol., 5, 914 (1991).
4.
Lane, W.F., and Loehr, R.C., Estimating the Equilibrium Aqueous

Concentrations of Polynuclear Aromatic Hydrocarbons in Complex
Mixtures, Environ. Sci. Technol., 26, 983 (1992).
5.
Rixey, W.G., Garg, S., and He, X., A Methodology for Accounting for the
Finite Leaching Characteristics of Contaminated Soils and Oily Wastes on
Groundwater Dilution and Attenuation, Environmental Engineering
Science, 17, 117-127 (2000).
172
Chapter 11
6.
nl, K., Kemblowski, M.W., Parker, J.C., Stevens, D., Chong, P.K., and
Kamil, I., A Screening Model for Effects of Land-Disposed Wastes on
Groundwater Quality, J. Contaminant Hydrology, 11, 27-49 (1992).
7.
van Genuchten, M.Th., and Alves, W.J., Analytical Solutions of OneDimensional Convective-Dispersive Solute Transport Equation, U.S.
Salinity Lab., Riverside, California, U.S. Department Agric.-Agric. Res.
Serv., Tech. Bull. No. 1661, 149 pp (1982).
8.
Millington, R.J., and Quirk, J.M., Permeability of Porous Solids, Trans.

Faraday Soc., 57, 1200-1207 (1961).
9.
NMED, Guidelines for Corrective Action, New Mexico Underground

Storage Tank Bureau, New Mexico Environmental Department, March 13,
2000.
10. Cussler, E.L., Diffusion: Mass Transfer in Fluid Systems, Cambridge

University Press, Cambridge (1984).
11. Carslaw, H.S., and Jaeger, J.C., Conduction of Heat in Solids, 2nd Ed.,
Oxford, Oxford University Press (1959).
12. DeVaull, G.E., Ettinger, R.A., Salanitro, J.P., and Gustafson, J.B.,
Benzene, Toluene, Ethylbenzene, and Xylenes [BTEX] Degradation in
Vadose Zone Soils During Vapor Transport: First-Order Rate Constants, in
the Proceedings of the 1997 Petroleum Hydrocarbons and Organic
Chemicals in Ground Water Conference, National Ground Water
Association, Westerville, Ohio, pp 365-379 (1997).
13. ASTM, Standard Guide for Risk-Based Corrective Action Applied at
Petroleum Release Sites, E-1739-95, American Society for Testing
Materials (1995).
14. ASTM, Standard Guide for Risk-Based Corrective Action, E-2081-00,
American Society for Testing Materials (2000).
15. Domenico, P.A., An Analytical Model for the Multidimensional Transport
of a Decaying Contaminant Species, Journal of Hydrology, 91, 49-58
(1987).
16. Brakensiek et al., Transactions of the American Society of Agricultural
Engineers, 335-339 (1981).
17. Carsel, R.F., and Parrish, R.S., Developing Joint Probability Distributions
of Soil Water Retention Characteristics, Water Resources Research, 24,
755-769 (1988).
173
Chapter 12
Risk Evaluation of Metals in Crude Oils

Renae I. Magaw,1 Sara J. McMillen,1 William R. Gala,1 John H. Trefry,2 and
Robert P. Trocine 2
1
2
Chevron Research and Technology Company;

Florida Institute of Technology
Introduction
Metals and semi-metals are commonly found in the environment (hereafter
metals and semi-metals are simply referred to as metals). They are present in
the rocks, soil, and organic matter that are the building blocks for the earth.
Some metals (such as chromium, selenium, and zinc) are essential to life and
must be supplemented as trace elements in the diets of humans and animals.
However, adverse health effects may be produced in people or other
environmental receptors when they are exposed to metals at certain
concentrations and under certain exposure conditions. For example, millions of
people inhabiting regions having iodine-deficient soils in eastern Africa are
susceptible to goiter, while Itai-Itai disease in China is attributed to people living
in areas where soils are contaminated by cadmium-containing wastes [1]. Some
metals , such as arsenic, cadmium, chromium, and nickel, have been shown to
produce cancer in people under some exposure conditions. Some metals have
been shown to produce adverse reproductive and/or other types of health effects
in people (e.g., lead, cadmium, mercury) and other animals (e.g., mercury,
selenium).
The chemical species of a metal is important in determining a metals toxicity.
For example, the inorganic form of arsenic is believed to be the carcinogenic
form, while the organic forms are not. Also, hexavalent chromium is
carcinogenic, while trivalent and elemental chromium are not. Therefore,
knowledge of the specific form of metal in an environmental sample (such as an
oil, waste, or soil sample) is important for accurate risk evaluation. Although it
is well known that different chemical forms of the same metal have different
toxicities, the analytical methodologies commonly employed do not readily
distinguish between different metal species. As a result, risk evaluations for
metals are usually based on the very conservative assumption that any metal
detected in a sample is in a form that may produce toxicity. This assumption
results in an overestimation of the potential risks posed by metals in
environmental samples.
Metals are natural components of crude oil. The chlorophyll molecule in
decomposing vegetative matter loses magnesium during crude oil formation, and
the magnesium is replaced by vanadium or nickel. The amount of vanadium and
174
Chapter 12
nickel in crude oils is well documented and their relative abundance can be used
to identify the source and age of crude oils [2]. However, there is little
published information on other metals in crude oils that may be of concern to
human health and the environment.
In contrast, the metals content of drilling muds has been investigated by the
United States Environmental Protection Agency (USEPA) [3]. Drilling muds
contain elevated levels of some specific metals such as barium. The American
Petroleum Institute (API) has also analyzed drilling muds and other exploration
and production (E&P) wastes (e.g., tank bottoms and produced sand) for metals
[4].
Through a joint industry research project carried out by the Petroleum
Environmental Research Forum, a more thorough understanding of the types
and concentrations of metals in crude oil was gained. The identification of
particular species of metals in crude oils was not attempted. Commonly
employed analytical methodologies were used, and it was then assumed that the
metals detected were in a form that may produce toxicity for risk evaluation
purposes, consistent with standard risk assessment practice. Therefore, as
discussed above, the risk evaluation for the metals detected in this study is
conservative. The technical information obtained in this study is needed by
regulators, risk assessors, and site managers to implement risk-based decisions
at sites that have been impacted by crude oil.
The Current Status of Metals Regulations & Risk Assessment

The State of Louisiana is currently the only state within the United States that
routinely requires the analysis of metals in E&P wastes. The La29B regulatory
limits for metals are provided in Table 1 [5]. These regulatory limits are not
risk-based values. In 1995, the API developed risk-based guidance levels for
metals to be used in the land management of E&P wastes based upon the
assumptions and calculations developed by the USEPA for land application of
sewage sludge [6,7]. Both API and USEPA evaluated 14 different exposure
pathways of concern for ecolo gical and human health. The APIs maximum soil
concentrations for 11 of the 18 metals analyzed in this study are shown in
Table 1. However, to date, no state regulatory agency in the United States has
adopted the API metals guidance as regulatory criteria. In the absence of
specific regulations for metals, comparison of metals concentrations to the API
criteria or to Louisianas regulatory levels may be useful.
In some cases, a risk assessment of metals has been required by local regulators
to determine whether metals in soil impacted by crude oil might pose a risk to
human health or the environment. This can be very difficult to do because it is
often impossible to conclusively differentiate metals that may be naturally
175
Chapter 12
present in soils from those that may have originated from a crude oil spill, or
from other sources.
The USEPA and many state agencies have developed human health risk-based
screening levels for metals. These screening levels are not regulatory limits, but
are commonly used as Tier 1 screening tools for evaluating risks, as described in
Chapter 1. The USEPA has developed Preliminary Remediation Goals (PRGs)
[8] for evaluating potential human health effects at Superfund sites. The PRGs
for metals are provided in Table 1. As an example of a state regulatory
program, the Texas Natural Resource Conservation Commission (TNRCC) has
developed Protective Concentration Levels (PCLs) under the Texas Risk
Reduction Program (TRRP) and the lowest residential PCLs for all exposure
pathways are also listed in Table 1 [9].
PRGs and PCLs represent chemical concentrations in environmental media (soil,
water, and air) that are considered protective of humans, including sensitive
groups, over a lifetime of exposure. They can be used to evaluate potential
health risks, trigger further site investigation, and serve as initial remediation
goals. Chemical concentrations above PRGs or PCLs do not automatically
mean that a site must be cleaned up to ensure public health, but they do suggest
that further evaluation of the potential risks posed by site contaminants may be
in order.
The PRGs presented in Table 1 are based on human exposure pathways
involving direct contact with contaminated soil and they focus on incidental
ingestion, dermal contact, and inhalation of vapors or soil particles containing
the chemicals of concern. They are derived by combining current USEPA
toxicity values with standard exposure factors and they correspond to fixed
levels of risk [i.e., either a one-in-one million (1 x 10-6) cancer risk or a noncarcinogenic hazard quotient of one], depending upon the risk levels selected by
the regulatory agency. The Texas PCLs are based on the same direct contact
exposure pathways, but they include additional indirect exposure pathways such
as potential leaching to groundwater and uptake into garden vegetables grown in
residential surface soil. They correspond to a target cancer risk level of one-inone-hundred thousand (1 x 10-5) and a non-carcinogenic hazard quotient of one.
For antimony, barium, beryllium, cadmium, chromium (total), lead, selenium,
and thallium, the lowest residential PCL is based on the metal leaching to
groundwater. PCLs for direct human contact with impacted soil are higher than
the levels presented.
176
Chapter 12
Table 1. Comparison of regulatory limits, PRGs, miscellaneous guidance

levels, and mean concentrations of metals in soils of the United States (U.S.).
USEPA Region IX
PRG (mg/kg)
Ag (Silver)
As (Arsenic)
Ba (Barium)
Be (Beryllium)
Cd (Cadmium)
Co (Cobalt)
Cr (Chromium
(Total))
Cu (Copper)
Hg (Mercury)
Mo (Molybdenum)
Ni (Nickel)
Pb (Lead)
Sb (Antimony)
Se (Selenium)
Sn (Tin)
Tl (Thallium)
V (Vanadium)
Zn (Zinc)
Texas
Lowest
Residential
PCL
Residential Industrial (mg/kg)
390
10,000
0.48
0.39
2.7
24
5,400
100,0007 440
API
Guidance
Level
(mg/kg)
NA
41
180,000
Lowest
Ecological
SSL6 (mg/kg)
2
ORNL-P
10 ORNL-P
500 ORNL-P
Mean
U.S. Soil
Conc.
[10]
(mg/kg)
ND
5.2
440
150
37
4,700
210 3
2,200
810
100,0007
450 3
1.8
1.5
1,300
2,400
NA
26
NA
15003
4
3
20
0.4
CCME-A
CCME-A
ORNL-P
ORNL-E
0.63
ND
6.7
37
NA
10
NA
NA
500
NA
10
NA
NA
2,900
23
390
1,600
400
31
390
47,000
5.2
76,000
610
10,000
41,000
750
820
10,000
100,0007
130
550
2.1
49
160
3
5.4
2.3
35,000
1.7
750
17
37 4
210
300
NA
100 5
NA
NA
60
0.1
2
30
50
5
1
2
1
17
0.058
0.59
13
16
0.48
0.26
.89
ND
NA
500
550
23,000
14,000
100,0007
290
2,400
NA
1400
2
50
ORNL-E
ORNL-E
ORNL-P
ORNL-P
ORNL-P
ORNL-P
ORNL-P
CCME-A
ORNL-P
CCME-A
ORNL-P
ORNL-P
La29B
(mg/kg)
NA
10
20,000 1a
40,000 1b
100,0001c
NA
10
NA
500
58
48
PRG = Preliminary Remediation Goal

PCL = Protective Concentration Level from Texas Risk Reduction Program
NA = Not Available
1) Louisiana 29B values for (a) wetlands, (b) uplands, and (c) commercial landfarming facilities.
2) A hot water soluble method for extraction of boron is recommended (Carter, 1993, Soil Sampling and Methods
of Analysis, Boca Raton, Lewis Publishers, pp 91-94). The guidance for boron is based on the soluble
concentration with units of mg/L rather than the total concentration (mg/kg).
3) The chromium value given in the USEPA PRGs is based on a combination of chromium (VI) and chromium
(III), while the API assumed that all chromium in soil would be chromium (III).
4) Under certain conditions this level of molybdenum may not be protective of grazing livestock. These
conditions are alkaline soils under arid and semi-arid conditions with deficient levels of copper in the soil.
5) The potential for plant uptake of selenium may be high in alkaline soils under arid or semi-arid conditions.
Plants that accumulate selenium in such soils may pose a threat to grazing animals. If elevated levels of
selenium are found in the waste, safeguards should be taken to limit this exposure pathway.
6) Sources of lowest ecological benchmarks are: ORNL-P = screening benchmark to protect terrestrial plants [12];
CCME-A = CCME remediation criteria for agricultural land-use [13]; ORNL-E = screening benchmark to
protect earthworms [11].
7) PRGs for relatively less toxic inorganic contaminants are set at a ceiling limit of 100,000 mg/kg. These values
are not risk -based and represent a USEPA Region IX policy decision. Health risk-based PRGs for these metals
would be higher than 100,000 mg/kg.
Ecological soil screening levels (SSLs), or benchmarks, have been developed by

groups in North America and Europe. Ecological SSLs are chemical
concentrations in soil below which it is unlikely that the chemical of potential
concern would pose an unacceptable risk to ecological receptors. These may be
considered Tier 1 screening levels to be used for evaluating potential risks to
environmental receptors. As with human health-based PRGs and PCLs,
chemical concentrations above ecological SSLs do not mean that a site must be
177
Chapter 12
cleaned up, but they do suggest that the chemical(s) should be further evaluated
in an ecological risk assessment.
Ecological SSLs may be specific to a particular type of ecological receptor such
as plants, invertebrates, or microbial communities, and they may be specific to a
particular type of land use. The lowest ecological benchmarks available from
published sources are shown in Table 1. Sources for ecological soil benchmarks
include the Oak Ridge National Laboratory (ORNL) soil benchmarks for soil
and litter invertebrates (e.g., earthworms), heterotrophic soil microbial
communities, and terrestrial plants [11,12]; Canadian Council of Ministers of the
Environment (CCME) remediation criteria for agricultural land-use [13]; and the
Dutch National Institute of Public Health and the Environment (RIVM)
Ecotoxicological Intervention values [14]. Note that ecological SSLs have not
been developed for specific hydrocarbons in soil (e.g., benzene and
naphthalene); however, some have been developed for hydrocarbon fractions of
a Canadian crude oil [15].
In any risk evaluation for metals, it is critical to consider the background levels
of metals in soils. A thorough investigation of the concentrations of metals in
uncontaminated background soils in the United States has been reported by the
United States Geological Survey [10]. The mean values reported for metals in
soils in the United States from this publication are also provided in Table 1.
Laboratory Procedures Used to

Measure Metal Concentrations in Crude Oils
Twenty-six crude oils were analyzed for 18 metals that may pose a risk to
human health or the environment. The metals concentrations of the crude oils
are provided in Table 2. Figure 1 illustrates the geographic locations from
which the crude oil samples were obtained. The metals contents were
determined by Florida Institute of Technology researchers, as described below.
178
Chapter 12
6
4
1
2 1
1
2
1
3
region from which they originated.
For all metals except mercury and selenium, a 1-gram aliquot of each crude oil
was weighed into a glass digestion flask and sealed with a glass watch cover.
The samples were moved to hotplates and then digested with concentrated, highpurity sulfuric acid (H2SO4), nitric acid (HNO3), and hydrogen peroxide. A
separate digestion was performed for selenium, with 1-gram subsamples of each
crude oil and high-purity HNO3 as the only oxidizing acid. For mercury
determinations, a 0.1-gram subsample of each crude oil was weighed into a glass
digestion tube and oxidized with concentrated, high-purity HNO3 and H2SO4.
Once the crude oil digestions were completed, the samples were placed in
graduated cylinders, diluted to 10 mL with reagent water rinses of the digestion
tubes, and stored for analysis in 15 mL polyethylene bottles.
Crude oil Standard Reference Materials (SRMs) with trace metal concentrations
certified at environmental levels do not exist; therefore, additional crude oil
subsamples were weighed and then spiked with the elements of interest. Acid
digestion was selected for crude oil decomposition rather than ashing techniques
for two reasons: (1) the lower temperatures used during acid digestion were less
likely to cause the loss of volatile elements, and (2) the risk of sample
contamination was less.
Metal concentrations of the digested crude oil samples, spiked samples, and
blanks were determined by one of four methods: flame atomic absorption
spectrometry (FAAS), graphite furnace atomic absorption spectrometry
(GFAAS), inductively coupled plasma-mass spectrometry (ICP-MS), or coldvapor atomic absorption spectrometry (CVAAS). Concentrations of zinc (Zn)
were measured by FAAS using a Perkin-Elmer Model 4000 AAS. Cadmium
179
Chapter 12
(Cd), cobalt (Co), chromium (Cr), copper (Cu), molybdenum (Mo), nickel (Ni),
and vanadium (V) concentrations were determined by GFAAS with a PerkinElmer Model 4000 AAS utilizing a HGA-400 graphite furnace and AS-40
autosampler. A Perkin-Elmer Model 5100 AAS with HGA-600 graphite
furnace and AS-60 autosampler was used to measure concentrations of silver
(Ag), arsenic (As), beryllium (Be), and selenium (Se) by GFAAS.
Concentrations of barium (Ba), lead (Pb), antimony (Sb), tin (Sn), and thallium
(Tl) were measured by ICP-MS using a Perkin-Elmer ELAN 5000 spectrometer.
Crude oil mercury (Hg) concentrations were determined by CVAAS using a
Laboratory Data Control Model 1235 Mercury Monitor.
Results: Metals Concentrations in Crude Oils

The concentrations of metals in each crude oil tested are presented in Table 2. A
summary of the data, including the method detection ilmits, is provided in
Table 3. The method detection limits for the metals were 6 to 1,000 times lower
than the suggested USEPA reporting limit for soils.
Table 2. Amount of metals in crude oils (data are in mg/kg oil).
Africa #1
Africa #2
Africa #3
Africa #4
Asia
Central America
Indonesia #1
Indonesia #2
Indonesia #3
Middle East #1
Middle East #1*
Middle East #2
North America #1
North America #2
North America #3
North America #4
North America #5
North America #5*
North America #6
North America #7
North America #8
North America #9
North America #10
North America #11
North Sea
South America #1
South America #2
South America #3
API
Gravity
35.0
37.1
37.8
33.3
46.4
20.9
19.4
32.3
31.9
32.5
32.5
33.5
29.3
36.1
40.8
39.2
22.6
22.6
40.7
28.1
13.3
43.4
30.7
26.0
19.5
12.0
19.2
16.2
Ag
0.07
0.11
0.15
0.09
0.11
0.07
0.16
0.14
0.23
0.14
0.14
0.16
0.17
0.30
0.10
0.11
0.13
0.15
0.20
0.15
0.28
0.21
0.19
0.09
0.05
0.08
0.14
0.18
As
0.19
ND
ND
ND
ND
0.17
ND
0.57
0.09
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.19
ND
0.23
ND
0.19
ND
ND
ND
Ba
ND
ND
ND
0.054
0.032
0.002
0.189
0.041
0.036
ND
ND
ND
0.003
0.018
ND
0.368
0.087
0.090
0.011
0.036
0.206
ND
0.006
0.014
0.124
0.015
ND
0.095
Be
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Cd
0.006
0.003
0.014
0.026
0.016
0.005
0.020
0.008
0.010
0.006
0.007
0.016
0.005
0.015
0.007
0.006
0.006
0.007
0.005
0.003
0.013
0.003
0.011
0.006
0.005
0.010
0.016
0.009
Co
0.44
ND
ND
0.38
ND
0.11
0.69
0.38
0.30
ND
ND
ND
0.02
0.02
0.02
ND
1.13
1.08
ND
ND
0.81
ND
ND
1.33
0.67
0.27
0.16
0.22
Cr
0.035
0.021
0.369
0.124
0.067
0.211
1.43
0.667
0.869
0.074
0.079
0.016
0.248
0.111
0.033
0.022
0.864
0.874
ND
0.167
0.398
0.095
0.329
0.379
0.069
0.117
0.088
0.214
Cu
0.031
0.023
0.058
0.083
0.069
0.031
0.052
0.147
0.098
0.024
0.028
0.046
0.048
0.133
0.050
0.079
0.173
0.165
0.059
0.046
0.148
0.012
0.241
0.234
0.055
0.068
0.026
0.071
Hg
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1.56
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
180
Chapter 12
Table 2 (Continued). Amount of metals in crude oils (data are in mg/kg oil).
Africa #1
Africa #2
Africa #3
Africa #4
Asia
Central America
Indonesia #1
Indonesia #2
Indonesia #3
Middle East #1
Middle East #1*
Middle East #2
North America #1
North America #2
North America #3
North America #4
North America #5
North America #5*
North America #6
North America #7
North America #8
North America #9
North America #10
North America #11
North Sea
South America #1
South America #2
South America #3
API
Gravity
35.0
37.1
37.8
33.3
46.4
20.9
19.4
32.3
31.9
32.5
32.5
33.5
29.3
36.1
40.8
39.2
22.6
22.6
40.7
28.1
13.3
43.4
30.7
26.0
19.5
12.0
19.2
16.2
Mo
0.49
0.50
0.60
0.43
0.58
4.01
0.85
0.41
0.41
0.47
0.45
0.87
0.41
0.31
0.35
0.48
0.54
0.53
0.62
0.31
0.71
0.41
0.53
1.91
0.78
1.29
0.71
0.97
Ni
3.20
0.33
0.87
7.28
0.08
53.5
24.1
9.39
4.62
4.32
4.48
4.32
12.3
4.30
5.09
2.28
50.4
51.8
0.30
14.1
55.9
0.05
9.97
57.8
6.87
93.0
36.4
50.3
Pb
0.008
0.010
0.014
0.018
0.009
0.008
0.025
0.028
0.015
0.009
0.011
0.025
0.024
0.101
0.005
0.045
0.006
0.005
0.035
0.038
0.111
0.005
0.069
0.018
0.018
0.022
0.020
0.149
Sb
ND
0.001
ND
ND
ND
0.036
0.010
0.005
0.010
0.008
0.011
0.019
0.003
0.002
ND
0.005
0.013
0.012
0.012
0.004
0.010
0.001
0.009
0.017
0.001
0.055
0.028
0.022
Se
0.05
0.03
N.D
0.02
0.04
0.23
0.04
0.04
0.03
0.08
0.08
0.14
0.12
0.27
0.06
0.04
0.52
0.52
0.05
0.12
0.44
N.D
0.13
0.31
0.13
0.46
0.24
0.43
Sn
2.72
0.11
0.81
1.50
9.66
1.23
3.26
0.44
1.30
0.90
0.91
2.41
2.33
0.42
0.31
2.43
0.27
0.27
0.54
0.11
0.11
0.14
0.49
2.26
0.04
0.08
0.18
1.45
Tl
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.004
ND
ND
0.004
ND
ND
ND
0.002
ND
ND
ND
ND
ND
V
0.13
0.47
1.3
0.91
1.2
320
1.4
1.2
0.15
21
21
7.2
33
20
1.0
4.6
45
44
0.44
40
100
0.36
0.66
120
42
370
250
250
Zn
4.06
2.30
3.23
2.50
4.81
0.58
0.63
6.04
1.28
2.61
2.61
4.21
2.30
3.70
8.42
3.96
3.54
3.70
2.04
2.89
0.59
ND
10.9
3.39
1.10
ND
0.66
ND
ND = Not Detected
* = Duplicate Analysis
181
Chapter 12
Table 3. Statistical summary of metals content of 26 crude oils (the data are in
mg/kg oil).
Mean
Minimum
Maximum
Detection Frequency
(# per 26 Oils)
Method Detection Level
Suggested USEPA
Reporting Limit
Mean
Minimum
Maximum
Detection Frequency
(# per 26 Oils)
Method Detection Level
Suggested USEPA
Reporting Limit
Ag
0.15
0.05
0.30
26
As
0.06
ND
0.57
7
Ba
0.052
ND
0.368
19
Be
ND
ND
ND
0
Cd
0.010
0.003
0.026
26
Co
0.27
ND
1.3
16
Cr
0.270
ND
1.43
25
Cu
0.081
0.012
0.241
26
Hg
0.06
ND
1.56
1
0.010
0.1
0.080
0.5
0.001
1
0.005
0.1
0.002
0.2
0.010
0.2
0.005
0.5
0.010
0.5
0.010
0.6
Mo
0.77
0.30
4.0
26
Ni
19.69
0.05
93.0
26
Pb
0.032
0.005
0.149
26
Sb
0.011
ND
0.055
21
Se
0.16
ND
0.52
24
Sn
1.37
0.04
9.66
26
Tl
0.000
ND
0.004
3
V
62.75
0.13
370.0
26
Zn
2.92
ND
10.9
23
0.020
0.2
0.020
1
0.001
1
0.001
1
0.02
1
0.010
---
0.002
1
0.020
0.5
0.080
1
ND = Not Detected
The mean values for the oils were less than 1.5 mg/kg for all metals except
nickel, vanadium, and zinc, which had mean values of 19.7, 62.8, and
2.9 mg/kg, respectively. Beryllium was not detected in any of the crude oils.
Only one oil from the San Joaquin Valley in the State of California contained
mercury at a concentration above the detection limit. Mercury in this oil has
been previously reported, and its presence is believed to be due to the proximity
of the oil reservoir to mercury (quicksilver) deposits [16]. Thallium was
detected in only 3 of the oils, and arsenic was detected in only 7 of the 26 crude
oils.
Evaluation of Human Health Risks

From Metals in Crude Oils
One way of evaluating the potential health risk associated with metals in crude
oils is to assume that crude oil has been spilled on soil and that people might
come into direct contact with the impacted soil. The potential health risk can
then be evaluated by comparing the concentrations of the metals in the impacted
soil to the screening levels presented in Table 1.
The concentrations of each metal in crude oil, as presented in Table 2, were
compared to the USEPA direct human contact PRGs and the TNRCC lowest
residential PCLs shown in Table 1. This comparison indicates that for 23 of the
26 crude oils, the metals concentrations in the oils did not exceed any PCL or
PRG for any metal. One of the crude oils contained arsenic at a concentration
182
Chapter 12
greater than its PRG, and two contained vanadium at concentrations greater than
its PCL. For the oil exceeding the arsenic PRG, the crude oil concentration in
soil that would result in an arsenic concentration exceeding the PRG would be
666,666 mg/kg, or 66.7% total petroleum hydrocarbons (TPH). Similarly, for
the oil containing the highest concentration of vanadium, the oil content in soil
would have to be more than 78.4% TPH in order to exceed the vanadium PCL in
Texas.
Therefore, it is highly unlikely that metals in soil from a single crude oil spill or
a single application of oil to the land surface would result in a significant risk to
human health. Sites that have received multiple applications of oil or oilcontaminated soils, such as a landfarm site, would require further study. Also,
E&P wastes may contain higher concentrations of some metals than their parent
crude oils due to corrosion processes, chemical additives, or metals in produced
water, and more data may be required to evaluate their risks.
Evaluation of Ecological Risks

From Metals in Crude Oils
The concentrations of metals in the 26 crude oils were compared to ecological
SSLs to evaluate the potential need for performing ecological risk assessments
at crude oil spill sites. As shown in Table 4, 12 metals do not exceed the lowest
ecological SSL even at their maximum concentration in pure oil. Only
chromium, mercury, molybdenum, nickel, tin, and vanadium are present at
levels that would limit the amount of crude oil in soil. For all metals except
vanadium, maximum crude oil concentrations of 60,000 mg/kg (6%) in soil
would result in maximum metals concentrations below their respective
ecolo gical SSLs.
183
Chapter 12
Table 4. Comparison of lowest ecological benchmarks with the maximum

concentrations found in 26 crude oils.
Chemical
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium
Cobalt
Copper
Lead
Mercury
Molybdenum
Nickel
Selenium
Silver
Thallium
Tin
Vanadium
Zinc
Lowest Ecological
Benchmark 1 (mg/kg)
5
ORNL-P
10
ORNL-P
500
ORNL-P
4
CCME-A
3
CCME-A
0.4
ORNL-E
20
ORNL-P
60
ORNL-E
50
ORNL-P
0.1
ORNL-E
2
ORNL-P
30
ORNL-P
1
ORNL-P
2
ORNL-P
1
ORNL-P
CCME-A
2
CCME-A
2
ORNL-P
50
ORNL-P
Maximum Concentration in
Crude Oil (mg/kg oil)
0.055
0.567
0.368
<0.005
0.026
1.43
1.33
0.241
0.149
1.56
4.01
93
0.52
0.296
0.004
9.66
370
10.9
1) Sources of lowest ecological benchmarks are: ORNL-P = screening

benchmark to protect terrestrial plants [12]; CCME-A = CCME remediation
criteria for agricultural land-use [13]; ORNL-E = screening benchmark to
protect earthworms [11].
Vanadium is the limiting metal for many of the crude oils. For the crude oil
containing the highest level of vanadium (370 mg/kg), the maximum
concentration of crude oil in soil that would not exceed the lowest vanadium
SSL is 5,400 mg/kg. A maximum concentration of 10,000 mg/kg (1%) crude oil
in soil would be protective for all oils, except for those from Central and South
America. For the North American crude oils analyzed (which contained as
much as 120 mg/kg vanadium), 16,000 mg/kg or 1.6% crude oil would not
exceed the lowest vanadium ecological SSL.
The lowest ecological SSL for vanadium is a benchmark derived by ORNL and
it is based on effects in terrestrial plants. It is possible that this benchmark is
overly conservative, since it is almost 30-times lower than the average
background concentration for vanadium in soil in the United States. Therefore,
it is unlikely that vanadium would pose a significant risk to ecological receptors
even at concentrations exceeding 1% crude oil in soil.
184
Chapter 12
Based on our comparison of metal concentrations in crude oils to ecological

SSLs, it is clear that there is no need to routinely perform ecological risk
assessments for metals at sites contaminated by crude oil spills and/or historic
releases. It may be necessary to perform a screening risk assessment to
determine if site-related metals could potentially be of ecological concern at
those sites for which other ecological pathways and receptors are important
(such as wildlife and aquatic organisms) or at sites that have received multiple
applications of oil.
Conclusions
Twenty-six crude oils were analyzed for 18 metals. Evaluation of the human
health risk associated with soil containing crude oil showed that the potential
risk was not significant at total oil concentrations in soil well above current
management levels of 10,000 mg/kg TPH, used in many states. The amount of
metals in 10,000 mg/kg TPH would also be protective of soil invertebrates,
plants, and soil microbial communities. Vanadium may be of ecological concern
only in some heavy crude oils from Central and South America.
It is apparent that acceptable levels for crude oil in soil based on the potential
human health effects of metals are well above those that would be expected to
produce unacceptable aesthetic effects. Overall, the concentrations of these
metals in crude oils are unlikely to be a major risk management consideration at
crude oil spill sites. Therefore, routine analyses for metals in soils at crude oil
spill sites is not recommended.
References
1.
Chang, L.W., Toxicology of Metals , Boca Raton, CRC Lewis Publishers,

(1996).
2.
Hunt, J.M., Petroleum Geochemistry and Ge ology, San Francisco, W.H.

Freeman & Company (1979).
3.
U.S. Environmental Protection Agency (USEPA), Technical Report Exploration, Development, and Production of Crude Oil and Natural Gas,
Field Sampling and Analysis Results, Office of Solid Waste and
Emergency Response, Publication 530-SW-87-005, Washington, D.C.
(1987).
4.
American Petroleum Institute (API), Oil and Gas Industry Exploration and
Production Wastes, API Document No. 471-01-09, Washington, D.C.
(1987).
5.
Louisiana Department of Natural Resources, Amendment to Statewide

Order 29-B, Office of Conservation (1990).
185
Chapter 12
6.
American Petroleum Institute, Metals Criteria for Land Management of

Exploration and Production Wastes: Technical Support Document for API
Recommended Guidance Values, API Publication No. 4600, Washington,
D.C. (1995).
7.
U.S. Environmental Protection Agency (USEPA), Standards for the Use or

Disposal of Sewage Sludge, 40 CFR Parts 257, 403, & 503 (1993).
8.
U.S. Environmental Protection Agency (USEPA) Region IX, PRGs Table

2000 Update, San Francisco, California, November 3, 2000.
9.
Texas Natural Resource Conservation Commission (TNRCC), Texas Risk

Reduction Program (TRRP) Rule, Updated Protective Concentration Level
(PCL) Tables, March 15, 2001.
10. Shacklette, H.T., and Boerngen, J.G., Element Concentrations in Soils and
Other Surficial Materials of the Conterminous United States, United States
Geological Survey Professional Paper 1270 (1984).
11. Efroymson, R.A., Will, M.E., and Suter, G.W., Toxicological Benchmarks
for Contaminants of Potential Concern for Effects on Soil and Litter
Invertebrates and Heterotrophic Process: 1997 Revision, ES/ER/TM126/R2, Oak Ridge National Laboratory, Oak Ridge, Tennessee (1997).
12. Efroymson, R.A., Will, M.E., Suter, G.W., and Wooten, A.C.,
Toxicological Benchmarks for Contaminants of Potential Concern for
Effects on Terrestrial Plants: 1997 Revision, ES/ER/TM-85/R3, Oak Ridge
National Laboratory, Oak Ridge, Tennessee (1997).
13. Canadian Council of Ministers of the Environment (CCME), Interim
Canadian Environmental Quality Criteria for Contaminated Sites, CCME
EPC-CS34, Winnipeg, Manitoba (1991).
14. van den Berg, R., Denneman, C.A.J., and Roels, J.M., Risk Assessment of
Contaminated Soil: Proposal for Adjusted, Toxicologically Based Dutch
Soil Clean-up Criteria, Arendt, F., Annokkee, G.J., Bosman, R., and van
den Brink, W.J. (eds), Contaminated Soils 93, Kluwer Academic
Publications, The Netherlands (1993).
Standards for Petroleum Hydrocarbons (PHCs) in Soil: Scientific Rationale,
Supporting Technical Document, December 2000.
16. Bailey, W.H., Snavely, Jr., P.D., and White, D.E., Chemical Analyses for
Brines and Crude Oil, Cymric Field, Kern County, California, Short
Papers in the Geologic and Hydrologic Sciences , Articles 293-435
Geological Survey Research 1961, Paper 424-D, pp D-306-309.
186
Chapter 13
Application of Risk-Based
Decision-Making for International
Exploration and Production Site Management
Sara J. McMillen,1 Renae I. Magaw,1 Michele Emerson,2 and Lynn R. Spence3
1

Brown and Caldwell
3
2
Introduction
In general, the fundamental process of risk-based decision-making (RBDM) is
understood and accepted in the United States. In fact, as discussed in previous
chapters, several states have incorporated RBDM into their site remediation
programs. Howe ver, for most countries outside of North America and Europe,
the concept of RBDM is relatively new or unfamiliar. A few international
agencies, such as the Canadian Council of Ministers of the Environment [1] and
the Dutch Directorate-General for Environmental Protection, Department of Soil
Protection [2] have recently proposed or adopted risk-based approaches to
derive regulatory levels for total petroleum hydrocarbons (TPH). Appropriately,
they have utilized the same health risk-based concepts as those used in the
United States as they are equally applicable to sites anywhere in the world.
Moreover, in developing specific action levels for its program, each agency has
incorporated specific assumptions that reflect conditions in its own part of the
world to ensure that they are protecting the health of their citizens.
However, in other areas of the world this is not the rule. Most regulatory
programs appear to more commonly set cleanup criteria based on a review of
levels that have been set by North American or European agencies, rather than
on a site- or even country-specific basis. The assumptions that were used by the
North American or European agencies in setting cleanup levels are rarely
reviewed to determine whether they are indeed appropriate for the new
application in a given country. It is incorrect to merely assume that United
States or European calculations, which reflect potential exposures in those parts
of the world, are necessarily appropriate for other parts of the world. In order to
develop meaningful regulatory programs and cleanup levels, adequate
consideration must be given to factors such as differences in lifestyles, climate,
geology, etc., which apply to different parts of the world. This chapter presents
two examples in which RBDM and risk-based screening levels (RBSLs) were
applied to exploration and production (E&P) sites in Nigeria and Indonesia.
These examples demonstrate some of the important factors that must be
considered in applying RBDM at international locations.
187
Chapter 13
Developing RBSLs for Nigerian Sites

In the following example, a generic conceptual site model for a crude oil
terminal in a Nigerian delta is developed. The focus of this example is on
human health considerations, not potential ecological impacts. This example is
not a comprehensive, site-specific risk assessment; rather it is simply an
example of how a RBDM program might be implemented at a Nigerian site [3].
Prior to applying RBDM at Nigerian sites, the Nigeria Department of Petroleum
Resources and other stakeholders would need to be consulted regarding proper
use of Nigeria-specific parameters and policy decisions.
Site Description
The generic site in this example is assumed to be a large crude oil terminal
located adjacent to a major river delta. Spills occurring at the site are assumed
to be contained within the site boundaries, and it is assumed that there are no
floating hydrocarbons on the water table. The soil at the site is sandy fill and
maximum depth to groundwater is 3 feet. However, shallow groundwater is
brackish and thus non-potable. Drinking water for the site is supplied by water
wells that draw from 1,800 to 2,000 feet below ground surface.
Employees live on the site in company housing constructed of concrete blocks.
The buildings have windows and are air conditioned. Workers are on duty for
7 days, then have 7 days off when they leave the terminal. The site is
adequately protected from trespassers by a fence and guards, so a trespasser risk
scenario does not need to be considered. Workers at the terminal wear
protective equipment including long pants and long-sleeved shirts, boots, and
hard hats.
There is residential housing just outside the fenced terminal. This housing is
also constructed of concrete blocks, but these are typically open-air buildings,
(i.e., not air conditioned). Subsistence fishing in the nearby river is the main
occupation. Residents may grow their own vegetables, but there is no large
scale farming or grazing nearby.
A conceptual site model (CSM) is provided in Figure 1 illustrating the exposure
pathways that may occur in case of a leak or spill of crude oil at the site.
Workers may be exposed via direct contact withsoil, incidental ingestion of soil,
or inhalation of vapors or soil particles from surface soil. Exposure may also
occur via inhalation of vapors from subsurface soil. It is unlikely that workers
would be exposed to hydrocarbons via drinking groundwater because the
shallow groundwater aquifer is not used as a potable water source, and it is
unlikely that hydrocarbons would migrate more than 1,800 feet, the depth of the
drinking water wells.
188
Chapter 13
Secondary
Source
Surface
Soils
Crude Oil
Terminal
Transport
Mechanism
Suspension
of Particulates
Subsurface
Soils
Volatilization
Groundwater
Infiltration,
Leaching
Surface
Water
Exposure
Media
Soil
Air
Outdoor
Groundwater
Surface
Water
Exposure
Pathway
Ingestion
Dermal contact
Inhalation
Inhalation
Receptor
On
sit
eW
ork
Of
er
fsit
eW
On
ork
er
sit
eR
es
id
Of
fsit ent
e
Re
sid
en
t
Primary
Source
X
X
X
Ingestion
Dermal Contact
Consumption
of Fish
= Potentially complete exposure pathway
Figure 1. Potential exposure pathways at a generic oil terminal site in Nigeria.

The nearby residents use rainwater as a drinking water supply. Therefore, it is
assumed that the local residents would only be exposed via inhalation of vapors
from surface or subsurface soils and inhalation of surface soil particles, as long
as the spill or leak was contained within the fenced terminal site. At this site,
consideration should also be given to the potential impacts to surface water from
hydrocarbon migration from the groundwater to the river. If hydrocarbons
migrate to the river, there is a potential human health exposure pathway related
to ingestion of fish caught in the river.
Exposure Parameters and Equations
The RBSL calculations followed the general approaches and algorithms used by
the American Society for Testing and Materials (ASTM) [4], the Total
Petroleum Hydrocarbon Criteria Working Group (TPHCWG) [5], and the
Petroleum Environmental Research Forum (PERF) [6]. The variables and the
specific exposure assumptions used in the analysis are presented in Tables 1 and
2. These are also consistent with those presented by ASTM [4]. However, a
few modifications were made to better reflect current risk assessment practices
and to better address Nigeria-specific parameters.
189
Chapter 13
Table 1. Receptor-specific parameter values used for the RBSL calculations.
Parameter
Lifetime
Body Weight
Exposure Duration
Exposure Frequency
Soil Ingestion Rate
Inhalation Rate
Soil to Skin Adherence Factor
Skin Surface Area Exposed to Soil
Units
Years
kg
Years
d/Year
mg/d
m 3/d
mg/cm2
cm2
Resident
Adult
52
70
24
365
NA
22
NA
NA
Resident
Child
52
15
6
365
NA
10
NA
NA
Worker
52
70
25
183
100
22
0.2
7,000
NA = Not applicable, because inhalation of vapors and soil particles is the only
exposure pathway.
Table 2. Media and source geometry parameter values used for the RBSL
calculations.
Parameter
Lower Depth of Surficial Soil Zone
Fraction Organic Carbon in Soil
Particulate Emission Rate
Wind Speed Above Ground Surface in Outdoor
Air Mixing Zone
Width of Source Area in Major Direction of Wind
Outdoor Air Mixing Zone Height
Volumetric Air Content in Vadose Zone Soils
Total Soil Porosity
Volumetric Water Content in Vadose Zone Soils
Soil Bulk Density
Averaging Time for Vapor Flux
Units
cm
g OC/g Soil
g/cm 2-s
cm/s
Value
100
0.002
2.3E-12
225
cm
cm
(cm 3 Air)/(cm 3 Soil)
(cm 3 Voids)/(cm 3 Soil)
(cm 3 H2O)/(cm 3 Soil)
g/cm 3
s
1500
200
0.26
0.38
0.12
1.7
= Exposure Duration
(see Table 1)
Two receptors were considered for the generic Nigerian crude oil terminal
described above: an onsite worker and a resident adjacent to the site. The
worker is assumed to be at the site half of the year (approximately 183 days).
For many workers in the United States , the exposure time (used in the inhalation
calculations) is assumed to equal 8 or 9 hours per day (hr/day). Because the
onsite workers in this example are at the site 24 hr/day, the exposure time was
adjusted to 24 hr/day. It is possible that the workers could come in direct
contact with residually impacted soil, i.e., soil containing oil that has been left in
place after emergency response cleanup has occurred. The RBSLs developed in
this paper are for long-term worker exposures to oil impacted soil. The
remediation worker who handles the spill initially, but only over a short period
of time, is not addressed in this analysis.
190
Chapter 13
The exposure pathways considered for the workers are incidental ingestion of
soil, dermal contact with soil, and inhalation of soil vapor and particulate
emissions. While on-duty, the workers wear full protective clothing and it is not
likely that incidental ingestion and dermal contact would actually occur. These
exposure pathways were included to take into consideration the times when
workers may be off-duty and wearing lighter clothing, playing games like
soccer, or doing other miscellaneous activities around the site. The parameter
values assumed for the workers are shown in Table 1. The average lifespan in
Nigeria was obtained from the international database of the United States
Census Bureau [7]. The soil ingestion rate of 100 mg (per day) is twice the
value normally used in the United States for industrial scenarios, but it seems
appropriate given that the Nigerian workers are at the site for a much longer day
than workers in the United States . The skin surface area of 7,000 cm2 assumes
that the worker typically wears shorts and a short-sleeved shirt when off-duty,
but still on site [8]. The inhalation rate of 22 m3/day is recommended by the
World Health Organization [9] and is similar to the value typically used in the
United States . [8]. Worker exposure due to inhalation of indoor air is not
included in this example because of the many site-specific parameters that must
be investigated prior to calculating RBSLs for this pathway.
The resident is assumed to live at the sites property boundary. Because spills
occurring at the site are assumed to be contained within the property boundary,
the resident will not come in direct contact with the oil. Vapors and soil
particulates covered with oil may be carried by the wind to the residences, and
therefore the exposure pathways considered for the resident will be the
inhalation of vapor and dust emissions. When calculating risk from noncarcinogenic chemicals, the degree of risk is calculated by comparing the
estimated average daily dose with an acceptable dose (the reference dose, or
RfD). For residential exposures to non-carcinogenic chemicals, it is standard
risk assessment practice to assume that the receptor is a child. Children
typically have high inhalation and ingestion rates relative to their body weight,
and therefore they are usually the receptor with the greatest potential risk
because they can experience the highest average daily dose. The exposure
parameters used in this analysis are for a child 1 to 6 years in age, and they are
shown in Table 1.
Nigeria Crude Oil RBSLs
Five crude oils from Nigeria were analyzed using the modified TPHCWG
analytical method (see Chapter 4). Their American Petroleum Institute (API)
gravities ranged from 24 to 38. The calculated TPH RBSLs for noncarcinogenic effects are shown in Table 3 and they are in units of mg -TPH per
kg-soil (mg/kg). For onsite workers, the RBSLs for surface soil exposure range
from 38,000 to 45,000 mg/kg. These levels of crude oil will not result in
191
Chapter 13
significant risk to adult workers who have direct contact with oil impacted soil.
For offsite residents, the surface soil RBSLs range from 51,000 to 62,000.
These levels of crude oil will not result in significant risk to children from
inhalation of petroleum vapors or oily soil particles. (Note that if a spill were
not contained within the terminal boundaries so that residents might be exposed
to surface soils, the RBSLs could be lower by at least a factor of 10.)
For onsite workers and offsite residents the subsurface soil RBSLs are RES.
The term RES, for residual saturation, is used to indicate that the oil does not
pose a significant risk of adverse health effects even at residual levels in soil.
(The RES concept applies only for the leaching to groundwater and volatile
exposure pathways.)
Table 3. TPH RBSLs for Nigerian oils at a terminal site.
Oil Location/
Source
Crude Oil #1
Crude Oil #2
Crude Oil #3
Crude Oil #4
Crude Oil #5
Surface Soil
Offsite
Onsite Worker
Resident Soil
Soil RBSL
RBSL (mg/kg)
(mg/kg)
51,000
38,000
56,000
43,000
60,000
45,000
57,000
44,000
62,000
41,000
Subsurface Soil
Offsite
Onsite
Resident Soil
Worker Soil
RBSL (mg/kg) RBSL (mg/kg)
RE S
RES
RES
RES
RES
RES
RES
RES
RES
RES
RBSLs were not calculated for the potential fish ingestion pathway identified in
the CSM. The only chemicals of potential concern in Nigerian crude oils that
could be taken up by fish and transferred to people consuming the fish are the
polyaromatic hydrocarbons. However, these chemicals are relatively insoluble
in water and would not be expected to be dissolved in groundwater and
transported to the river as a result of an onsite spill.
Developing RBSLs for Indonesian Sites

This section describes example TPH RBSLs developed for two different
exposure conditions relevant to E&P operations in Sumatra, Indonesia:
unrestricted land use and indemnified areas (non-residential). The TPH
RBSLs correspond to TPH levels in soil that should pose no significant risk to
human health with consideration given to oil composition, the current and future
land use, populations likely to be exposed (children versus industrial workers),
and exposure mitigation procedures.
There are two concerns to consider when setting cleanup levels or RBSLs in
Sumatra human health and ecosystem health.
192
Chapter 13
The natural ecosystem in Sumatra has been disturbed in some areas due to
logging, palm plantations, farming, and industrial activities. For these areas, it
is most appropriate to consider the risk to human health along with current and
future land use when setting cleanup levels or RBSLs. Therefore, RBSLs were
derived for specific oil types and considering specific land use (e.g., unrestricted
versus industrial or indemnified areas).
The science behind assessing risk to human health is well developed and has
been described at length in this book. However, data on the impact of
hydrocarbons in soil to specific ecological receptors (plants, animals, and
insects) within an ecosystem are very limited, and the data that are available are
largely from studies conducted with North American or European species.
Therefore, it is impossible at this time to set acceptable TPH limits for
protecting ecological receptors in undisturbed tropical rainforest ecosystems.
Site Description
Some oil production areas in Sumatra have been restricted to oilfield operations
only and are termed indemnified or restricted use. These areas can therefore
be evaluated much as industrial or commercial sites are evaluated in the United
States. However, oil production may also co-exist with other land uses such as
farming, residential areas, or plantations. These land uses were evaluated as
unrestricted or similar to United States residential land use scenarios.
Figure 2 illustrates a CSM for an unrestricted land use in Sumatra.
Primary
Source
Secondary
Source
Surface
Soils
Crude Oil
Production
Sites
Subsurface
Soils
Groundwater
Transport
Mechanism
Exposure
Media
Exposure
Pathway
Receptor
Onsite
Resident
Suspension
of Particulates
Volatilization
Infiltration,
Leaching
Surface
Water
Soil
Air
Outdoor
Groundwater
Surface
Water
Ingestion
Dermal contact
Inhalation
Inhalation
X
X
X
Ingestion
Dermal Contact
X
X
Ingestion
X = Potentially complete exposure
Figure 2. Potential exposure pathways at crude oil production sites in Sumatra,

Indonesia for unrestricted land use.
193
Chapter 13
Application of TPHCWG Method to Sumatran Crude Oils

Two main types of crude oils are produced in Sumatra: Sumatran Light (Minas
crude oil is one example of this type oil), which has an API gravity of 32, and
Duri, which has an API gravity of 19.4. These crude oils were analyzed by the
modified TPHCWG analytical method and the results are shown below in
Figure 3. Duri crude oil has a higher concentration of aromatic hydrocarbons
and vacuum residuum (>C44) than Minas crude oil, since the Duri oil has been
biodegraded in the reservoir. Sumatran Light oils like Minas are waxy crude
oils and have high levels of large molecular weight alkanes .
Duri
Minas
500000
450000
TPH, mg/kg oil
400000
350000
300000
250000
200000
150000
100000
50000
>C
44
>6
-8
Ali
ph
atic
>8
s
-10
Ali
ph
ati
>1
cs
0-1
2A
liph
atic
>1
s
2-1
6A
liph
ati
>1
cs
6-4
4A
liph
ati
cs
>7
-8
Ar
om
atic
>8
s
-10
Ar
om
>1
ati
cs
0-1
2A
rom
atic
>1
s
2-1
6A
rom
atic
>1
6-2
s
1A
rom
atic
>2
s
1-4
4A
rom
atic
s
TPH Equivalent Carbon Fractions
Figure 3. The TPH fractions [aliphatic and aromatic] and vacuum residuum
>C44 fraction for Minas and Duri oils.
RBSL Calculations for Crude Oils in Soil
The exposure assumptions commonly used in the United States were reviewed
for their suitability for deriving TPH RBSLs to be used in Indonesia. Very little
published information is available for developing Indonesia-specific exposure
parameters. A few exposure parameters were modified based on World Health
Organization information or site specific measurements s(uch as the fraction
organic carbon in soil). These are described in Table 4.
194
Chapter 13
Table 4.
RBSLs.
Parameter
Body Weight
Indonesia-specific parameters used to calculate non-cancer TPH
Abbrev. Units
BW
kg
Ingestion Rate: IRsoil

Soil
200
50
300
100
Inhalation Rate: IRair-out m3/Day

Air-Outdoor
10
20
15
22
Ingestion Rate: IRw

Water
L/Day
1.5
Skin Surface
Area
SA
cm2/Day 2,900
3,160
3,900
4,100
Exposure
Frequency
EF
Days/
Year
350
250
365
365
g/g
0.01
0.01
0.002
0.002
Fraction Organic Foc

Carbon in Soil
mg/Day
United States
Sumatra Indonesia
Residential Comm./ Residential Comm./
(Child)
Indust. (Child)
Indust.
15
70
12
60
Rationale for Sumatra

Value
Average body weight as
described by World
Health Organization for
adult workers [9]. Body
weight for child assumed
to be proportional to
difference between U nited
States and Sumatra adult
body weights [8].
United States
recommended soil
ingestion rate for people
involved in activities with
heavy soil contact [8].
World Health
Organization
recommendation for
commercial/industrial
scenario [9].
World Health
Organization
recommendation for
adults in environments
with high average
temperatures [9].
Recommendation for
child is 50% more than
value assumed for United
States child [8].
Skin surface area for
Sumatra commercial/
industrial scenario
corresponds to a person
wearing shorts, shoes and
a short-sleeved shirt [8].
Surface area for Sumatran
child corresponds to a
person wearing shorts and
a sleeveless shirt [8].
Conservative assumption
based on best professional
judgment.
Measured value for total
organic carbon content of
Indonesian soils.
RBSLs for Restricted and Unrestricted Land Use

RBSLs were developed for oily soil using an unrestricted land use scenario. For
this situation, RBSLs of 1,500 mg/kg TPH were calculated for Duri and 2,500
mg/kg TPH for Sumatra Light-type crude oils/wastes. Unrestricted use includes
such high exposure activities as living (children and adults) on the site soil,
constructing ponds, and using groundwater underneath the site as the sole water
195
Chapter 13
supply. These RBSLs apply to TPH concentrations in surface soils, roads, oil
spills, or oily pit sites that are located on or near places where people live.
The RBSLs for unrestricted land use are low due to the living conditions in a
largely rural, developing region like Sumatra. The limiting pathways are dermal
exposure, soil ingestion, and soil particle inhalation. Leaching to groundwater
and volatile emissions will not usually be pathways of concern due to the low
concentratio ns of benzene (<50 mg/kg oil) and other low molecular weight
aromatic hydrocarbons in both Duri and Sumatra Light crude oils.
For restricted land use the RBSLs are much higher. The TPH RBSL for Duri
crude oil is 18,000 mg/kg TPH and is 30,000 mg/kg TPH for Sumatra Lighttype crude oils . This is due to lower amounts of exposure for industrial oil field
workers.
Conclusions
The RBDM process is a scientifically defensible, flexible, yet standard process
that can be used to develop international RBSLs. Simply adopting cleanup
standards that have been developed by European or North American regulatory
agencies for conditions in their countries will not provide meaningful standards
for developing countries.
At a generic terminal site in Nigeria, a RBDM approach is used to determine
potential receptors and complete exposure pathways. In the example presented,
TPH RBSLs are needed for three commercial worker pathways and two
residential exposure pathways. Similarly, example RBSLs were developed for
use in Sumatra Indonesia based on two land use scenarios. For both countries,
exposure parameters were modified to be protective of the way people live and
may be exposed to chemicals in their environments. In cases where site
conditions exceed these RBSLs, they can be used by a site manager as cleanup
levels or a more detailed Tier 2 or 3 risk assessment could be performed. It may
also be possible to reduce risk to human health by taking actions to eliminate
some of the pathways of concern.
References
1.
Canadian Council of Ministers of the Environment (CCME), A Protocol

for the Derivation of Environmental and Human Health Soil Quality
Guidelines, CCME-EPC-101E (1996).
2.
National Institute of Public Health and Environmental Protection (RIVM),

Human Exposure to Soil Contamination: A Qualitative and Quantitative
Analysis Towards Proposals for Human Toxicological Intervention
Values, Report No. 725201011, Bilthoven, The Netherlands, January
1994.
196
Chapter 13
3.
McMillen, S.J., Magaw, R.I., and Spence, L.R., Developing Risk-Based

Cleanup Standards for Soils in Nigeria, SPE Paper 66513, presented at the
SPE/EPA/DOE Environmental Conference, San Antonio, Texas (2001).
4.
ASTM, Standard Guide for Corrective Action Applied at Petroleum

Release Sites, ASTM E 1739-95, American Society for Testing and
Materials, West Conshohocken, Pennsylvania (1995).
5.
Vorhees, D., Gustafson, J., and Weisman, W., Human Health Risk-Based
Evaluation of Petroleum Contaminated Sites: Implementation of the
Working Group Approach, TPHCWG, Vol. 5, Amherst Scientific
Publishers, Amherst, Massachusetts (1999).
6.
McMillen, S.J., Magaw, R.I., Kerr, J.M., Sweeney, R.E., Nakles, D.V., and
Geiger, S.C., A New Risk-Based Approach to Establish Cleanup Levels
for Total Petroleum Hydrocarbons, presented at the International
Petroleum Environmental Conference, Houston, Texas, November 1999.
7.
United States Census Bureau (USCB), International Data Base, Year 2000
Average Life Expectancy for Both Sexes in Nigeria (2000).
8.
U.S. Environmental Protection Agency, Exposure Factors Handbook

(1997).
9.
World Health Organization, International Programme on Chemical Safety,

Assessing Human Health Risks of Chemicals: Derivation of Guidance
Values for Health-Based Exposure Limits, Environmental Health Criteria,
170, 54 (1994).
197
Appendix
The Total Petroleum Hydrocarbon Criteria

Working Group (TPHCWG) Analytical
Method: Characterization of C6 to C35
Petroleum Hydrocarbons in Environmental
Samples
Total Petroleum Hydrocarbons
Alipha tic Hydrocarbons
Aromatic Hydrocarbons
Approximate Boiling Point/Carbon
Distribution
1.0
Number
Scope and Applications

1.1
This gas chromatographic method is designed to determine the

concentrations in soil and water of petroleum hydrocarbons
from n-hexa ne (C6) to n-pentatriacontane (C35); an
approximate boiling point range from 70C to 500C. This
includes the gasoline, diesel range, and some portions of
heavier fuels and lubricating oils. This method also describes
the separation of the petroleum hydrocarbons into their
aliphatic and aromatic fractions.
1.2
This method describes the characterization of the total

petroleum hydrocarbons, the aliphatic, and the aromatic
fractions into approximate carbon number/boiling ranges with
respect to n-alkane markers. See Figure 1 for overall
method options and when to apply them.
1.3
This method can be used to measure concentrations of

individual target analytes. When target analyte information is
desired, quantitation should be performed from the aliphatic or
aromatic fractions rather than from the unfractionated extract.
This will minimize the error due to coelution problems.
However, target analytes are best determined using EPA
Methods 8021, 8260, or 8270, where appropriate [1].
1.4
This method uses flame ionization (FID) as the mode of

detection. The response of the FID is generally equal for all
198
Appendix
hydrocarbons on a weight and effective carbon number basis

[2].
1.5
The method reporting limit is estimated to be 50 mg/kg in soil

and 5 mg/L in water depending on the number of hydrocarbon
components present in the C6 to C35 range. A limited
interlaboratory evaluation of the method for total petroleum
hydrocarbons in soil and in water has been conducted in Texas
to validate TNRCC TX Method 1005 [3]. In addition, a
previous version of this method (using a split mode of
injection) for total petroleum hydrocarbons in soil was
subjected to an interlaboratory study conducted by The
American Petroleum Institute [4]. It was found to have a PQL
from 50 to 104 mg/kg (depending on the definition of PQL),
an average accuracy of 84%, an average single analyst relative
standard deviation (RSD) of 13%, and an average overall RSD
of 30%. A similar study performed on this approach and the
fractionation procedure by a single analyst showed an average
accuracy of 80% with an average overall RSD of 6%. Also,
an independent laboratory evaluation of this method resulted
in a single analyst average accuracy of 111% and an overall
RSD of 10%. Additional evaluation of this method has been
done by the American Association of Railroads for
applicability for diesel range materials [5] and by A.D. Little,
Inc., for applicability to crude oil impacted soil. The latter
effort studied the efficiency of n-pentane as an extraction
solvent compared to methylene chloride and of vortex mixing
versus Soxhlet. Both solvents and extraction mechanisms
were found to be equivalent.
1.6
Petroleum and petroleum products with the majority of

hydrocarbon components in the 70C to 500C boiling point
range can be accurately extracted and measured by this
method. This range includes gasoline, kerosene, Diesel/Fuel
Oil No. 2, some lubricating oils, and portions of other heavier
oils.
1.7
This method should be used by, or under the supervision of,

analysts experienced in the use of solvent extraction, solid
phase fractionation, and gas chromatography. The analysts
should also be skilled in the interpretation of capillary gas
chromatography data (specifically petroleum hydrocarbon
pattern recognition), quantitation using computerized data
acquisition, and use of peak processing software with baseline
and peak grouping functions.
199
Appendix
1.8
This method was originally developed to characterize

petroleum hydrocarbons for proper remediation technology of
impacted s oils. Separation of the petroleum hydrocarbons into
an aliphatic and an aromatic fraction was developed to provide
data in the appropriate format to support the Total Petroleum
Hydrocarbons Criteria Working Group (TPHCWG) risk-based
corrective action approach to waste site remediation. This
approach is based on the fate and transport and toxicological
properties of petroleum hydrocarbon compound classes [6].
1.9
The extraction and fractionation procedure can take as little as

15 minutes to perform per sample. GC analyses may take 20
to less than 90 minutes depending on the chromatographic
column used and the GC parameters. Three separate GC
analyses per sample are required to obtain total petroleum
hydrocarbons, total aliphatics, and total aromatic information.
It is recommended that the sample extract be analyzed first to
determine the type of petroleum hydrocarbons (if any) in the
sample before proceeding with the fractionation step.
Additionally, this information can be used for potential source
identification, to assess if there are different types or
distributions of petroleum hydrocarbons in a sample, or to
determine if fractionation is necessary. If required, all or
some of the sample extract may be fractionated into aliphatic
and aromatic fractions, which are then analyzed by GC.
200
Appendix
Figure 1: Overall Method Options for Characterization of

Petroleum Hydrocarbons
GC/FID for
TPH
Extract Sample With

n-Pentane 1:1
No
GC Fingerprint
Similar for all
Samples?
No
2.0
Report
Results
Fractionate All
Samples
Fractionate by Alumina or Silica

Column Separation
GC/FID of npentane eluate

for Aliphatics
TPH Below
Action Level
Criteria?
Yes
Yes
Fractionate
Representative
Sample
Quantitate
TPH and BP
Distribution
Quantitate
Aliphatic and
Aromatic
Fractions
GC/FID of methylene
chloride (Alumina) or 1:1
methylene
chloride:Acetone (Silica)
eluate for Aromatics
Summary of Method
2.1
This method involves extraction of a soil or a water sample

with n-pentane and analysis of a portion of the extract using
gas chromatography with a flame ionization detector (GC-
201
Appendix
FID). For additional characterization, fractionation of the

petroleum hydrocarbon extract is accomplished by solid phase
separation of another portion of the extract using alumina
(similar to EPA Method 3611 [7]) and eluting with n-pentane
to obtain an aliphatic fraction followed by elution with
methylene chloride to obtain an aromatic fraction.
Alternatively, fractionation may also be done using silica gel
(similar to EPA Method 3630C [8]). Silica gel may be more
suitable for samples with a wide boiling point distribution of
hydrocarbons. Silica gel may also be better for the
fractionation of the higher molecular weight polynuclear
aromatics (PNAs). In the silica gel procedure, a 1:1 mixture
of acetone:methylene chloride is used to elute the aromatic
compounds. Other fractionation procedures, such as
automated HPLC methods, may also be used. The fractions
are also analyzed using GC-FID. The extract as well as the
fractions can be further characterized by subdividing the
chromatographic data into approximate boiling point/carbon
number ranges with respect to n-alkane markers.
2.2
This method allows choices of standards for calibration.

Either mixtures of single hydrocarbon components, petroleum
products (such as gasoline or diesel), or mixtures of petroleum
products can be used. It is strongly encouraged that petroleum
products similar to those present as contaminants in the
samples be used if possible.
2.3
This method is based in part on USEPA Methods 8000, 8015,

and 8100, SW-846, Test Methods for Evaluation of Solid
Waste, 3rd Edition [1]. It is also similar to the Massachusetts
Department of Environmental Protection Method for the
Determination of Extractable Petroleum Hydrocarbons (EPH)
[9]. This method is similar (for total petroleum hydrocarbons)
to Washington State WTPH-HCID [10]. It was developed at
Shell Development Company by I.A.L. Rhodes, L.P. Brzuzy,
et al. [11-13]. This method was the basis for TNRCC TX
Method 1005 [3].
2.4
This method uses n-pentane for the extraction of soil and

water samples.
Spiking studies done during method
development and subsequent experiments at several
laboratories with spiked and field samples show that n-pentane
is equivalent to methylene chloride in extraction efficiency of
hydrocarbons. The soil types ranged from sand to loam to
clay. In addition, the vortexing extraction as well as Soxhlet
202
Appendix
extractions were found to be equivalent (unpublished PERF

project results with crude oil in soil samples). Recent
published work by the University of Toronto indicates that npentane is an excellent solvent for extraction of hydrocarbons
from water [14].
3.0
Definitions
3.1
Total Petroleum Hydrocarbons (TPH) are defined as all gas

chromatographic peaks eluting after the solvent (n-pentane)
starting with and including n-hexane (nC6) to
n-pentatriacontane (nC35). This definition includes aliphatic
and aromatic hydrocarbons. The petroleum hydrocarbons in a
sample (if any) may not encompass the entire range. If the
range of compounds present is narrower, then it is best to
report on the observed range only. This information is useful
for product or source identification. There may be nonhydrocarbon compounds that elute in this range (such as
chlorinated solvents, ketones, alcohols, etc.). However, such
compounds usually appear as discrete peaks and do not match
typical petroleum product fingerprints. In some cases, such as
when the samples contain crude or motor oil, only the portion
within the nC6 to nC35 will be measured as TPH.
3.2
Aliphatic Hydrocarbons are defined as those compounds

detected from n-hexane (nC6) to n-pentatriacontane (nC35)
(inclusive) in the chromatogram of the aliphatic fraction.
3.3
Aromatic Hydrocarbons are defined as those compounds

detected from n-hexane (nC6) to the retention time of npentatriacontane (nC35) in the chromatogram of the aromatic
fraction. The first aromatic compound is benzene.
3.4
Approximate Boiling Point/Carbon Number Distribution

is defined as the subdivision of the chromatogram into
sections that correspond to boiling point and/or volatility of nalkanes. The gas chromatographic separation is achieved
using a column that separates components based primarily on
boiling point differences. This separation can be correlated to
approximate carbon number. For example, >C7 to C8
indicates those hydrocarbons that elute after n-heptane and up
to and including n-octane. This range includes most, but not
all, of the C8 hydrocarbons. Branching lowers the boiling
points of hydrocarbons relative to their n-alkane isomers.
Cyclization, or ring structures, raises the boiling point higher
203
Appendix
than the n-alkanes of the same carbon number. Thus, there are
some C8 hydrocarbons that elute before n-heptane and there
are some that elute after n-octane, including the aromatics
ethylbenzene and the xylenes.
This method allows for data reporting between each carbon
range or for reporting within wider carbon ranges depending
on data quality objectives. The TPHCWG has defined
fractions based on different properties that affect the fate and
transport and/or toxicity of petroleum hydrocarbon
components.
4.0
3.5
Aliphatic Hydrocarbon Standard may be used to calibrate

the analysis of the aliphatic or n-pentane fraction. The
standard may be prepared from a mixture of n-alkanes and
branched alkanes in n-pentane. This standard is not a
requirement since it is recommended that a petroleum product
or mixed products be used as standards for a single calibration
that can be applied to extract and fractions.
3.6
Aromatic Hydrocarbon Standard may be used as an option

to calibrate the analysis of the aromatic fraction. The standard
can be prepared from a mixture of benzene, toluene,
ethylbenzene, xylenes, C3-benzenes (such as npropylbenzene), C4-benzenes (such as t-butylbenzene), and
polynuclear aromatics (such as the EPA target PNAs)
prepared in methylene chloride. As indicated in 3.5, this
standard is not required.
3.7
Locator Mix Standard will be used to determine the ranges

C6 to C35 and the individual ranges specified in Table 2.
3.8
An Analytical Batch is defined as a set of 1 to 20 samples

prepared on the same day.
Interferences
4.1
Other organic compounds, including vegetable and/or animal

oils and greases, organic acids, chlorinated hydrocarbons,
phenols, and phthalate esters are measurable under the
conditions of this method.
However, if present, the
characteristic petroleum hydrocarbon patterns will be altered.
These compounds will be quantified as part of the TPH, but
the data should be flagged as presumptively containing a
significant amount of such compounds. The aliphatic and
204
Appendix
aromatic fractions may have less susceptibility to interferences

from some types of materials because the fractionation process
may remove the interference.
5.0
4.2
Sample contamination due to sample preparation may be

minimized by the use of disposable glassware. A reagent
blank should be analyzed with each set of 10 or less samples
to demonstrate that the system is free from contamination. If
samples are expected to have high concentrations, it is also
advised that solvent blanks be analyzed between GC runs to
minimize contamination due to carryover.
4.3
High purity reagent grade or pesticide grade n-pentane,

methylene chloride and acetone should be used to minimize
contamination problems.
4.4
This method depends on correctly integrating a mass of

unresolved peaks using a forced baseline. The resulting
baseline, if drawn incorrectly, will have a significant effect on
the concentration reported.
It is imperative that
chromatograms be checked (using a realistic scale relative to
the chromatogram) for correct baseline extension. Blanks
and/or a low level standard should be run to monitor for
baseline drift every 10 samples.
Health and Safety Issues

5.1
The toxicity of the reagents used in this method has not been
precisely defined. However, each chemical compound should
be treated as a potential health hazard. Exposure to these
chemicals should be reduced to the lowest possible level by
whatever means available. The laboratory is responsible for
maintaining a current awareness file of Occupational Safety
and Health Administration (OSHA) regulations regarding the
safe handling of the chemicals specified in this method. A
reference file or Material Safety Data Sheets (MSDS) should
also be made available to all personnel involved in the
chemical analysis. Additional references to laboratory safety
should be available and should be identified for use by the
analyst.
205
Appendix
6.0
Apparatus
6.1
Glassware
6.1.1
All specifications are suggestions only.
6.1.2
4 oz. (120 mL) amber glass wide-mouth jars.
6.1.3
Vials
6.1.3.1 10 to 40 mL glass vials with Teflon-lined
screw caps.
6.1.3.2 2 mL GC autosampler vials with Teflonlined crimp caps.
6.1.4
Disposable Pipettes: Pasteur.
6.1.5
1 cm I.D. by 10 to 20 cm glass column with glass or

Teflon stopcock.
6.1.6
Volumetric flasks and graduated cylinders.
6.2
Microsyringes: 10 L to 1000 L.
6.3
Analytical balance capable of accurately weighing 0.0001 g

should be used for preparation of standards. A top-loading
balance capable of weighing to the nearest 0.01 g should be
used for obtaining sample weights.
6.4
Vortex mixer.
6.5
Wrist action or horizontal shakers may be used for extraction.
6.6
Drying oven.
6.7
Gas Chromatography
6.7.1
Gas Chromatograph: Analytical system which

includes a splitless injector, column supplies, gases,
and syringes. Electronic Pressure Control (EPC) is
strongly recommended. A data system capable of
storing and reintegrating chromatographic data and
determining peak areas using a forced baseline and
baseline projection is required. A gas chromatograph
206
Appendix
capable of performing baseline compensation is

desirable.
6.7.2
Recommended Columns
6.7.2.1 25/30 m x 0.25 to 0.53 mm ID fused silica
capillary column with 0.25 to 1.5 m film
thickness (methyl silicone) or equivalent.
Low bleed columns are preferred. Examples
include MS-007 (Quadrex), DB-1 (J&W),
and RTX-1 (Restek). DB-5 (J&W) may be
used.
6.7.2.2 Other columns may be used if the elution of
the compounds is based on boiling point.
Capillary columns are recommended. See
Section 9.3.2 for GC performance criteria.
7.0
6.7.3
Detector: A flame ionization detector (FID) is

required.
6.7.4
Autosampler: An autosampler capable of making

1-4 L sample injections is recommended.
Reagents and Standards

7.1
n-Pentane, methylene chloride, acetone.

pesticide grade or equivalent.
Reagent grade,
7.2
Sodium Sulfate (ACS): Granular, anhydrous.

heating at 100C for 4 hours in a shallow tray.
7.3
Alumina, basic or neutral, Brockman activity 1, 150 mesh.

Activate by heating at 350C at least 12 hours before using.
Store at 110C until ready to use. Alternatively, silica gel, 75250 mesh. Activate at 110-130C until ready to use. In
addition, solid phase extraction cartridges, or automated
HPLC methods may be used but equivalency must be
demonstrated.
7.4
Calibration & Stock Standard Solutions: This method allows

for the choice of calibration standard for quantitation. The use
of either a petroleum product(s) standard or a standard
composed of selected hydrocarbons is acceptable. The
Purify by
207
Appendix
selected hydrocarbon standard is required for the definition of

carbon # retention windows and as a fractionation check
solution. Unless noted, standards are prepared in the npentane listed in 7.1 above. Standard preparation should
follow the guidelines outlined in EPA SW -846 Method
8000B.
7.4.1
Petroleum Product Calibration Standard for Total

Petroleum Hydrocarbons, Aliphatic and Aromatic
Fractions: The petroleum hydrocarbon calibration
standard can be prepared by accurately weighing
approximately 0.05 to 0.1 g (recorded to the nearest
0.0001 g) of a mixture of gasoline and Diesel #2 in a
1:1 (either by volume or weight) ratio and diluting to
volume with n-pentane in a 10 mL volumetric flask.
If only the gasoline range or the diesel range TPH is
of interest, then the calibration standards should be
prepared with either 0.100 g gasoline or 0.100 g
diesel. This 1% standard should be kept refrigerated.
Typical working concentration ranges are between 5
to 5000 g/mL. Calibration standards may be
prepared from a blend of selected hydrocarbons (as in
Section 7.4.2).
7.4.2
Petroleum Hydrocarbon Calibration Standard,

Approximate Boiling Point /Carbon Number
Distribution Marker, and Fractionation Check Stock
Standard: This standard can be used for several
purposes: for TPH calibration, for retention time
window/approximate boiling point distribution
marker (locator mix standard), and for alumina/silica
gel fractionation performance check. The stock
standard can be prepared by accurately weighing
approximately 0.01 g (recorded to the nearest
0.0001 g) of each of n-alkanes [n-hexane (C6)
through n-eicosane (C20) as well as n-pentacosane
(C25), n-octacosane (C28), and n-pentatriacontane
(C35)] and diluting to volume with n-pentane in a 50
mL volumetric flask. It is also suggested that this
standard contain 0.01 g each of benzene, toluene,
ethylbenzene, o,m,p-xylene, cumene, as well as some
or all of the target PNAs (naphthalene, anthracene,
pyrene, etc.). Table 1 lists the boiling points of the nalkanes. The laboratory should determine the
retention times. The approximate concentration of
208
Appendix
this stock solution is 200 g/mL per component. If

this solution is to be used as a marker for retention
time window/approximate boiling point distribution,
accurate concentrations are not necessary. If the
application requires that wider carbon ranges be used
(ex: C6 to C10), this standard can be prepared with
fewer n-alkane markers. This stock solution can be
used for the determination of total petroleum
hydrocarbons as well as the aliphatic and aromatic
fractions if a component standard is preferred over a
mixed product standard as described in 7.4.1 (please
note limitations of this approach listed in 9.5.1). If
available, a standard mix can be obtained from
commercial suppliers.
7.4.3
8.0
Petroleum Products Reference Standards: To assist in

the qualitative determination of product type or
fingerprint of a possible petroleum product(s), it is
recommended that a library of chromatograms be
generated of gasoline, kerosene, diesel, motor oil,
crude oils, and any other pertinent product for
comparison purposes. A recommended concentration
range is 1000 to 5000 ppm. These may be obtained
from several chromatography supply vendors.
Sample Collection, Preservation,

Containers, and Holding Times
8.1
Soil samples are collected in wide-mouth glass jars with

Teflon-lined caps. Soils samples can also be collected and
transported in core sampling devices [15]. Samples are stored
at 4C from the time of collection until extraction. Soil sample
extraction and analysis should be performed within 14 days of
collection. Depending on the analytes of interest and data
quality objectives, other holding times may be applicable.
8.2
Water samples are to be collected by filling a 40 mL (volatile

organics analysis) VOA vial with the sample and capping the
vial with a Teflon septum cap (headspace free). Water samples
may be preserved with HCl to a pH <2. Water sample
analysis must be performed within 7 days of collection.
209
Appendix
Table 1. Boiling points of n-alkanes used for the determination of approximate

boiling point/carbon number distribution. Retention times based on GC
conditions described in this method must be determined experimentally.
n-Alkane
Marker
n-C6
n-C7
n-C8
n-C9
n-C10
n-C11
n-C12
n-C13
n-C14
n-C15
n-C16
n-C17
n-C18
n-C19
n-C20
9.0
~ Boiling
Point, C
69
98
126
151
174
196
216
236
253
270
287
302
316
329
343
n-Alkane
Marker
n-C21
n-C22
n-C23
n-C24
n-C25
n-C26
n-C27
n-C28
n-C29
n-C30
n-C31
n-C32
n-C33
n-C34
n-C35
~ Boiling
Point, C
357
369
380
391
402
412
422
431
441
450
458
467
474
481
499
Procedures
9.1
Sample Extraction
9.1.1
Soil Extraction: Extract soil samples using a vortex

mixer or shaker technique.
9.1.1.1 Weigh 10 g of sample in a 40 mL vial with
Teflon cap. Record the weight to the nearest
0.01 g. If needed, add enough sodium
sulfate to make a loose-friable mixture (the
use of sodium sulfate may not be necessary
for dry soils). The sample should be free
flowing prior to addition of the n-pentane. It
is preferred that mixing of sodium sulfate
with the sample be done as quickly as
possible to minimize potential losses of
volatiles. Add 10 mL of n-pentane, cap the
vial, and proceed with the extraction.
210
Appendix
9.1.1.2 For a method blank, weigh 10 grams of

Ottawa sand or other blank standard soil and
extract as a sample.
9.1.1.3 For laboratory control samples (LCS) weigh
10 grams of Ottawa sand or other blank
standard soil. For matrix spikes, weigh a
separate 10-gram portion of sample. Add
0.25 mL to 1.0 mL of stock solution to both
the LCS and matrix spike as described in
Section 7.4.1 and extract them like the
samples.
9.1.1.4 Extract blanks , samples, LCS, matrix spike,
and matrix spike duplicates by vortexing for
at least 1 minute or shaking on a wrist action
or horizontal shaker for at least 1 hour.
9.1.1.5 If particulate is suspended in the solvent
layer or an emulsion forms, centrifugation
may be necessary to obtain a clear solvent
layer. Transfer a portion of the extract to
autosampler vials for direct analysis of the
extract for total petroleum hydrocarbons.
The extract may be stored in vials with
Teflon caps. Extracts should be stored at
-15C.
9.1.1.6 Anhydrous sodium sulfate may be used to
aid in the drying and extraction of wet
sediment or sludge samples. Weigh the
sample into the vial then add up to 10 g of
anhydrous sodium sulfate, cap and mix by
vortexing. Add 10 mL of n-pentane, cap the
vial, and proceed with the extraction.
9.1.1.7 If a sample of neat petroleum product, crude
oil, or waste is to be analyzed, the sample
should be diluted in n-pentane (1:50 to
1:100) and analyzed directly. Alternatively,
approximately 0.01 g (~1 drop) of the
material can be placed directly on the
column for fractionation (Section 9.2).
211
Appendix
9.1.2
Water Extraction: Extract water samples using a

vortex mixer.
9.1.2.1 Remove sample in a 40 mL VOA vial from
refrigeration and allow to come to ambient
temperature. Remove approximately 10 mL
of sample through the septum with a
syringe. It is recommended that a needle be
inserted into the septum at the same time to
allow for flow of air into the vial as the 10
mL of water are removed. Dry the outside of
the vial with a paper towel. Weigh the vial
and its contents on a top loading balance and
record the weight to the nearest 0.01 g.
9.1.2.2 Using a 5-mL glass syringe, add 3 mL of npentane through the septum of the 40 mL
VOA vial.
9.1.2.3 For laboratory method blanks, perform the
procedure as in Section 9.1.2.1 using a VOA
vial filled with approximately 30 mL of
reagent water.
9.1.2.4 For matrix spikes and laboratory control
samples (LCS), perform the procedure as in
Section 9.1.2.1, but add 0.1 mL of stock
from Section 7.4.1 to 30 mL of a sample or
distilled water prior to extraction.
9.1.2.5 Extract samples, LCS, blanks,
matrix spikes, and matrix spike
duplicates by vortexing for at least
1 minute.
9.1.2.6 Remove ext ract by pipette and
store in Teflon capped vials at 4C.
9.1.2.7 Discard the water from the VOA
vial and dry the vial, lid, and
septum in a drying oven at 70C.
9.1.2.8 Reassemble the vial and weigh it
on a top loading balance and record
the weight to the nearest 0.01 g.
212
Appendix
9.1.2.9 Determine the sample volume by

subtracting the initial vial weight
obtained in 9.1.2.1 from the dry
vial weight obtained in 9.1.2.8.
9.2.
Extract Fractionation: Fractionate extract into aliphatic and

aromatic components, if required, to obtain information in
format suitable for the risk based correction approach
proposed by the TPHCWG [6].
9.2.1
Prepare the column by placing approximately 1 cm of

moderately packed glass wool at the bottom of the
column. Assemble the stopcock making sure that it
turns smoothly.
9.2.2
Fill the column with about 10 mL of methylene

chloride. Add approximately 4 grams of activated
alumina to the column (or if silica gel is used, 2
grams of activated silica gel). Ensure that it is
packed uniformly by gently tapping the side of the
column. Top the column with approximately 0.5 cm
of sodium sulfate. Then rinse the column with at
least 10 additional mL of methylene chloride. Let the
solvent flow through the column until the head of the
liquid in the column is just above the top of the
column (alumina packing nearly exposed). Discard
the eluted methylene chloride. Add about 2 mL of npentane. Open the stopcock and let the solvent flow
until the liquid in the column is just above the top of
the column. Add 10-20 mL of n-pentane in the same
manner just described. Open the stopcock and let the
n-pentane flow until the head of the liquid is just
above the top of the column. Discard the eluant. The
column is ready for use.
NOTE: The performance of the alumina or the
silica gel is dependent on the particular lot
number of alumina or silica gel from the
manufacturer, the humidity of the laboratory
environment, and the activation temperature.
Each laboratory may need to raise or lower the
activation temperature depending on their
particular conditions to achieve optimal
separation.
213
Appendix
9.2.3
Add 1 mL of the sample extract to the column. Open

the stopcock and start collecting the eluant
immediately in a 10-mL graduated cylinder, a 10-mL
volumetric flask, or any appropriate measuring vial.
When the head of the n-pentane extract nearly
reaches the top of the alumina or silica gel column,
add n-pentane to the column in 1-2 mL increments
while continuing to collect the eluant. It is best to
add the solvent nearly dropwise with a pipette or
wash bottle. Continue this approach until an
accurately measured volume (8-10 mL) of the eluant
is collected. Cap the vial and label this fraction
aliphatics .
9.2.4
Once the 8-10 mL of the n-pentane (aliphatic)

fraction has been collected, proceed to collect in
another graduated cylinder, volumetric flask, or
appropriate measuring vial the aromatic fraction by
elution with methylene chloride. This is done in the
same manner as in 9.2.3 by collection of the eluant
immediately after addition of methylene chloride in
1-2 mL increments or dropwise. It is critical that the
first 3-4 mL be added carefully and slowly. Once
8-10 mL have been collected, cap the vial and label
this fraction aromatics . If silica gel is used, elute
this fraction with a 1:1 mixture of acetone:methylene
chloride.
9.2.5
Fractionation of neat petroleum products, crude oil,

and wastes is done by directly placing on the alumina
or silica gel column 1 drop of the sample or by
weighing approximately 0.01 g of the sample, adding
1 mL of n-pentane and then proceeding with the
fractionation as defined in Section 9.2.3.
NOTE: It is critical that extreme care be taken on
the elution of aliphatic and aromatic fractions to
optimize the fractionation process.
This
optimization can be achieved by allowing the
extract to elute from the column as much as
possible before the addition of additional solvent:
run the sample nearly out of the column before
more solvent is added. Add additional solvent in
214
Appendix
small increments to the column to separate and

obtain the fractions in narrow bands.
The amount of pentane and methylene chloride or
acetone:methylene chloride used to elute the
aliphatic and aromatic fractions, respectively, can
be optimized experimentally. Use enough pentane
to elute all the saturates. This may require as
little as 8 mL, but no more that 12 mL of pentane.
Recoveries of aliphatics should be greater than
80%. If more than 12 mL of pentane is used,
elution of aromatic compounds in the aliphatic
fraction may result. For the aromatic fraction,
use enough methylene chloride to ensure that all
the aromatic compounds, especially the PNAs,
have eluted from the column. Again, recoveries
should be greater than 80%. Minimizing the
amount of solvent used will increase the sensitivity
of the analysis by avoiding overdilution of the
sample.
9.3
9.2.6
Extract concentrations exceeding 10,000 g/mL TPH

may need to be diluted to avoid alumina column
overloading. Silica gel capacity has not been
determined; it is recommended that extraction
concentrations not exceed 10,000 g/mL.
9.2.7
The blank, LCS, matrix spike, and matrix spike

duplicate must also be fractionated with the sample
batch.
Gas Chromatography
9.3.1
Gas Chromatographic Conditions

9.3.1.1 Oven Program: Set the initial column oven
temperature to 10C and hold for 1 to
5 minutes. Then ramp at 4 to 10C/minute
to 320C and hold for 10 to 15 minutes.
Alternatively, set initial column oven
temperature to 30C and hold for 3 to 4
minutes. Ramp at 10 to 20C/minute to 300
to 320C, hold for up to 10 minutes. Any
oven program used must demonstrate
adequate separation between the solvent and
215
Appendix
n-hexane (the first compound to be included

in the TPH measurement).
9.3.1.2 Sample/Autosampler Injection: 1 to 4 L
splitless injection.
9.3.1.3 Carrier Gas: Helium at 15 psig for a
0.25 mm ID column or as recommended by
column manufacturer (lower pressures are
required for larger IDs). If electronic
pressure control (EPC) is used, the pressure
will be variable.
9.3.1.4 Make-up Gas: Nitrogen preferred (helium
can be used (30 mL/min.).
9.3.1.5 FID hydrogen and air set to manufacturers
specifications.
9.3.1.6 FID Temperature: 325C to 350C.
9.3.1.7 Injection Port Temperature: 300C
325C.
to
9.3.1.8 GC operated in splitless mode. Turn split on

1 minute after injection. Alternatively, a
direct injection (Uniliner) technique may be
used.
9.3.2
Performance Criteria: GC run conditions and

columns should be chosen to produce chromatograms
with adequate separation between the solvent front
and n-hexane (the earliest eluting compound of
interest that defines the beginning of the first carbon
range). In addition, clear baseline resolution in the
C6 to C12 range should be achieved.
NOTE: Adequate separation of n-hexane from
the solvent front (n-pentane) may be difficult with
thin film columns (<0.32 m film thickness). The
thin film columns facilitate the elution of
hydrocarbons up to n-C35 within a reasonable
time. Thus, there may be some columns that may
compromise the ability to analyze the entire C6 to
216
Appendix
C35 range adequately. The thicker film columns

work quite well in the C6 to C28 range.
9.4
Retention Time Windows

9.4.1
Before establishing windows for integration using nalkanes as markers, make sure that the GC system is
within optimum operating conditions. Make three
injections of the boiling point distribution standard
defined in 7.4.2 throughout the course of a 72-hour
period.
9.4.2
Calculate the standard deviation of the absolute

retention times for each individual component in the
boiling point distribution standard.
9.4.3
The width of the retention time window for each

analyte is defined as plus or minus three standard
deviations of the mean absolute retention time
established during the 72-hour period. If a standard
deviation of 0.00 is obtained, check reference
SW-846 8000B Section 7.6.3 for further instructions
[1]. Analyst experience should be part of the
interpretation of the chromatograms.
9.4.4
Alternatively, a default window may be chosen. This

approach is preferred over the one described above
because capillary columns are reliable with sufficient
overall long-term stability to maintain retention time
(RT) appropriately. This approach is also extremely
simple. A window of 0.1 minutes should be
adequate.
9.4.5
The laboratory should reassess retention time

windows for each standard on each GC column and
whenever a new GC column is installed.
9.4.6
TPH RT ranges are defined as beginning 0.1 minute

before the RT of the beginning marker compound (nhexane) and ending 0.1 minute after the RT of the
ending marker compound (n -pentatriacontane or the
last peak that elutes for a given sample if the last
peak elutes earlier than n-pentatriacontane). This RT
range is applicable for the determination of TPH,
aliphatic TPH, and aromatic TPH.
217
Appendix
9.5
9.4.7
The approximate boiling point/carbon number

distribution for the samples can be determined using
n-alkane to define carbon # ranges of interest. The
chromatograms obtained from analysis of TPH,
aliphatic TPH, and aromatic TPH can be subdivided
into individual boiling ranges/approximate carbon
numbers based on these RT windows.
The
TPHCWG has defined several aliphatic and aromatic
fractions for the RBCA approach. These fractions
encompass several carbon ranges which can be
obtained the same way but using either addition of
the individual carbon ranges or by defining wider
carbon ranges for measurements.
The marker
compounds are defined in Table 1.
9.4.8
TPHCWG Aliphatic and Aromatic Fractions: The

TPHCWG has defined approximate carbon ranges of
aliphatic and aromatic fractions based on fate and
transport considerations. The fractions are listed in
Table 2. This method can provide this information
by either defining the RT ranges using the
corresponding
n-alkane
markers
for
the
characterization of the aliphatic and aromatic
fractions with the exception of the aromatic fraction
where practical quantitation stops at C35.
Calibration: Calibrate the GC system using the external

standard procedure
9.5.1
The method takes advantage of the fact that the

response of the FID is essentially the same for all
hydrocarbons (on a weight basis) and based primarily
on effective carbon number [2].
Any other
compound containing heteroatoms will have some
reduced response with respect to hydrocarbons
because of lower carbon to hydrogen ratio. It is
therefore not essential that calibration be performed
using material similar to the material in the samples.
For example, any gasoline, diesel, synthetic mixture,
or single hydrocarbon can be used for calibration.
However, because products such as gasoline or diesel
are composed of more than 300 individual
components, at low concentration of total product,
many of the individual components are simply too
218
Appendix
small to be detected and cannot contribute to the total

signal detected and thus linearity falls off.
Conversely, when synthetic standards are used,
typically no more than 10-20 components are used
and thus the TPH is distributed among a few peaks
that can be all detected for all concentrations of the
standards above the stated practical quantitation
limits. The use of synthetic standards can result in
underestimation of the TPH present in the samples.
Table 2. TPHCWG fate and transport fractions.
Aliphatics
C6 to C8
>C8 to C10
>C10 to C12
>C12 to C16
>C16 to C21
9.5.2
Aromatics
>C6 to C7 (Benzene)
>C7 to C8 (Toluene)
>C8 to C10
>C10 to C12
>C12 to C16
>C16 to C21
>C21 to C35
External Standard Calibration Procedure: Guidance is

provided for calibration and calculations using
calibration factors.
However, it is strongly
recommended that a chromatographic data system be
used for data acquisition and processing. The
baseline may rise as a result of column bleed at the
higher temperatures towards the end of the run.
Baseline compensation may help in integration of the
chromatogram over the background from column
bleed.
9.5.2.1 Prepare calibration standards fro m one of
the stock solutions described in 7.4.1 or
7.4.2 at a minimum of five concentration
levels by adding volumes of the pertinent
stock standard solutions to volumetric flasks
and diluting to volume with n-pentane or
methylene chloride.
The following
calibration levels are recommended: 20, 50,
100, 200, 500, 1000 g/mL. Calculate the
total
concentration
of
the
7.4.2
multicomponent stock standard for TPH
219
Appendix
calibration by adding all of the

concentrations of the individual compound
(ex: 200 g/mL per component, if 20
components, then TPH is 4,000 g/mL)
before diluting to prepare the calibration
standards.
9.5.2.2 Inject each calibration standard using the
same injection volume (1 to 4 L injections)
and technique that will be used to introduce
the actual samples into the gas
chromatogram.
Tabulate peak area
responses against the concentration injected
using a force baseline projection. The
results can be used to prepare a calibration
curve for quantitation. Linear and quadratic
calibration fits may be adequate for the
calculation of sample results. Alternatively,
the ratio of the response to the amount
injected, defined as the calibration factor
(CF), can be calculated for the standard at
each concentration. If the percent relative
standard deviation (% RSD) is less than or
equal to 25% over the working range, the
average response factor can be used in place
of a calibration curve. If linear regression
analysis is used for quantitation, the
correlation coefficient (R) must be at least
0.995.
Calibratio n Factor (CF) =
Total Area of Calibratio n Standard

Concentrat ion of Calibratio n Standard ( g/mL)
Note: It is recommended that area

response from calibration standards be
acquired in the same manner as samples.
% RSD =
StandardDeviation of 5 CFs
x 100
Mean of 5 CFs
9.5.2.3 The calibration factor obtained in Section

9.5.2.2 can be used to calculate TPH,
aliphatic hydrocarbons, and aromatic
hydrocarbons. The same calibration factor
220
Appendix
can be used to calculate the concentration of

hydrocarbons present within the n-alkane
markers used for defining the approximate
boiling point/carbon number distribution.
Alternatively, the boiling point distribution
can be obtained from normalization of the
entire chromatogram (nC6 to nC35) and
determination of area percent within nalkane markers. It is best to use a
chromatographic data system to handle these
calculations. A separate calibration file with
the additional retention times should be
established for the hydrocarbon ranges of
interest to determine the approximate
boiling point distribution and/or selected
hydrocarbon ranges, both total and
fractionated.
9.5.2.4 The working calibration factor or calibration
curve should be verified on each working
day by the injection of a mid-point
calibration standard. If the concentration or
response for these standards varies from the
standard value or predicted response by
more than 25%, a new calibration curve
should be prepared. It is advisable to check
instrument performance and reanalyze a low
concentration standard as well to verify
instrument performance and linearity.
Relative Percent Difference =
R1 - R2
x 100
Ravg
where:
R1 = Standard value or average CF
R2 = Calculated value or CF
Ravg= (R1 + R2)/2
9.5.2.5 Calibration of Selected Target Analytes:
Selected components (target compounds
such as benzene, toluene, ethylbenzene,
xylenes, PNAs, n-alkanes) can be measured
individually if desired.
Assuming an
221
Appendix
equivalent calibration factor, the calibration

curve or calibration factor developed above
can be used for target analytes. Coelutions
of these target compounds in complex
hydrocarbon mixtures can be expected. This
is mo re significant for the PNAs because of
the large number of isomers that are present
as the carbon number increases. The results
from these analyses can result in
overestimation of these target compounds.
If necessary, confirmation and more
accurate quantitation may be obtained by
using EPA Methods 8021, 8260, 8270 [1].
9.5.2.6 Chromatographic Data System: The
concentration of specific analytes or
hydrocarbon ranges may also be calculated
from a calibration curve by use of regression
analysis.
9.6
Product Type Identification

9.6.1
Chromatographic peaks with characteristic fuel

fingerprints eluting between the solvent front and C10
indicate the presence of gasoline range. Peaks
between C10 and C25 indicate the presence of diesel
range compounds. Patterns that do not resemble
either product should be noted.
9.6.2
Product type can be determined by visual inspection

of the chromatograms. Chromatograms can become
more complicated if crude oil, jet range material, or
other refined products are also present. However, it
may still be possible to determine that the
contamination is due to some sort of petroleum
product. Industrial solvents can interfere in the
analysis; however, the chromatographic fingerprints
would be noticeably different. The best approach to
maximize the probability of a correct identification is
to analyze reference fuels, from the sample location,
along with the sample (if available). These reference
fuels can also be used as calibration standards if
desired.
222
Appendix
9.6.3
As with any gas chromatographic procedure using

non-selective
flame
ionization
detection,
interferences are possible from coelution of gasoline
components with soil contaminants of other sources.
Potentially, any compound with similar boiling point
and polarity as the hydrocarbons of gasoline-to-diesel
range may have retention times within the range of
interest and may result in overestimation of the TPH
concentration.
For example, volatile industrial
solvents, cleaners, and naturally occurring
compounds not of petroleum origin may interfere
with this analysis. It is often possible to recognize
the presence of solvents and cleaners since the
characteristic fingerprint of gasoline, kerosene,
diesel, and heavier materials is altered.
9.6.4
Decisions should be made by the analyst in

determination of cutoff points for quantitation of
different product ranges when contamination is
caused by a combination of sources. For example, if
soils are contaminated with gasoline range and diesel
range materials, there is an area of overlap where
certain components are common to both types of
petroleum fractions. A compromise cutoff for
mixtures of gasoline with diesel fuel range material is
C10. There is no appropriate cutoff for a mixture of
jet fuel or kerosene since there is a great deal of
overlap. Crude oil contamination also contains a
wide range of materials. In cases where mixed
products are present, it is perhaps best not to
quantitate how much is due to what type of product
but to simply quantitate total hydrocarbons and state
the approximate carbon range observed.
In order to minimize quantitation problems due to
column bleed, the method is best suited for analysis
of materials up to diesel range. Heavier materials can
be detected with a qualitative identification of
product mix but not quantitated effectively.
9.6.5
Additional information on hydrocarbon pattern

interpretation is included in some of the references
cited [11-13].
223
Appendix
9.7
9.8
Gas Chromatographic Analysis

9.7.1
Samples are analyzed by GC/FID.

Suggested
injection volumes are 1 to 4 L using the conditions
established in 9.3.
9.7.2
After initial calibration (9.5.2) has been performed,

verify the calibration by analysis of a mid-point
standard at the start of a new analytical sequence
using the criteria in 9.5.2.4.
9.7.3
For samples that contain unresolved hydrocarbons,

baseline projection should be used to generate the
area for TPH calculation or for a group within a
defined boiling range/carbon number. The GC
conditions used for this method produce minimal
column bleed up to C35.
9.7.4
Alternatively, if peak resolution is adequate,

valley-to-valley integration may be used to generate
peak areas. This is only possible in the gasoline
range (up to C12). The analyst should avoid
discarding chromatographic area related to
unresolved hydrocarbons.
9.7.5
If the product concentration exceeds the linear range

of the method in the final extract, the extract should
be diluted and reanalyzed. The upper end of the
linear range is defined as the highest standard in the
calibration curve. The linear range tested is
approximately equivalent to 50 g/mL to 10,000
g/mL of petroleum hydrocarbons in the extract.
Linearity beyond this range should be verified.
Calculations
9.8.1
External Standard Calibration: The concentration of

TPH, aliphatic hydrocarbons, aromatic hydrocarbons,
selected hydrocarbon ranges, or target analytes in a
sample can be calculated from the appropriate area
using either calibration factors or regression analysis.
Cs =
Cc x Vt x D
1 mg
x
Ws
1000 g
224
Appendix
where:
Cs = Concentration of TPH, hydrocarbon range or specific
analytes (mg/kg or mg/L).
Cc = Concentration from calibration curve in g/mL. (If CF is
used for calculations, this value is area calibration/CF).
Vt = Volume of extract (mL).
D = Dilution factor, if dilution was performed on the sample
prior to analysis. If no dilution was made, D = 1,
dimensionless.
Ws= Weight of sample extracted (kg). If a water sample, then
the units are L.
9.8.2
9.9
10.0
The peak areas may be divided into desired carbon

ranges/boiling point distribution if so desired.
Patterns that do not resemble hydrocarbon products
should be noted if the analyst is familiar with pattern
recognition/fingerprints of petroleum products.
Calculation of Approximate Boiling Point Distribution: The

approximate boiling point distribution is calculated by
normalization of sums of peak areas of portions of the
chromatograms eluting between preselected retention times as
indicated in Table 1. Actual retention times should be
verified in the laboratory. These retention times correspond to
known boiling points selected as references. Characterization
by individual approximate carbon number ranges is done up to
C35. This is only a guideline. Other markers or groupings can
be used. The chromatographic column used in this method is
primarily a boiling point, non-polar column. Compound
separation is achieved by boiling point differences. A
homologous series of n-alkanes is used to approximate boiling
point references.
Quality Control
10.1
General Requirements and Recommendations

10.1.1
The laboratory should establish the ability to generate

acceptable accuracy and precision. This should
include the analysis of QC check samples plus the
calculation of average recovery and the standard
deviation of the recovery as outlined in EPA Method
8000B, Section 8.0 [1].
225
Appendix
10.1.2
The laboratory should, on an ongoing basis,

demonstrate through the analysis of quality control
check standards that the operation of the
measurement system is in control. This should
include calibration verification every 10 samples,
method blank, LCS, matrix spike, and matrix spike
duplicate every 20 samples.
10.1.3
After successful calibration (Section 9.5), analyze a

reagent blank sample with every analytical batch or
sequence. The blank should not have petroleum
hydrocarbons above the practical quantitation limit.
In addition, n-pentane or methylene chloride solvent
blanks should be run after samples suspected of being
highly contaminated to determine if sample carryover
has occurred.
10.1.4
Each laboratory should generate control limits based

on the average recovery 3 standard deviations. For
the LCS, the laboratory must meet the minimum
criteria of 60-140% recovery for the whole TPH.
10.1.5
If any of the criteria in 10.3 and 10.4 are not met, the
problem should be corrected before samples are
analyzed.
10.1.6
Field blanks, duplicates, and matrix spikes are

recommended for specific sampling programs.
Matrix spikes should use the spike levels specified
for laboratory control samples.
10.1.7
Performance evaluation (PE) samples from an

independent commercial source for both soil and
water samples at both low (5-8 mg/L for water and
50-100 mg/kg for soil) and high (20-50 mg/L for
water and 1,000-20,000 mg/kg for soil) levels should
be analyzed prior to performing analysis. Data and
chromatograms for these PE samples must be kept on
file at the laboratory for audit purposes. The
performance evaluation samples should be analyzed
yearly.
226
Appendix
10.2
10.3
Minimum Instrument QC
10.2.1
The instrument should be able to achieve adequate

separation and resolution of peaks and analytes of
interest.
10.2.2
The n-hexane (nC6) peak should be adequately

resolved from the solvent in the chromatographic run.
10.2.3
Retention time windows should be established for

each analyte and/or carbon range of interest each
time a new GC column is installed, and should be
verified and/or adjusted on a daily basis.
10.2.4
Calibration curves should be developed based upon

the analysis of calibration standards prepared at a
minimum of five concentration levels. The linearity
of calibration or calibration factors may be assumed
if the percent relative standard deviation (% RSD)
over the working range of the curve is less than or
equal to 25%. Alternatively, if linear regression
analysis is used for quantitation, the correlation
coefficient (r) should be at least 0.995.
10.2.5
In order to demonstrate the absence of mass

discrimination, the response ratio of C35 to C20 should
be at least 0.80.
Initial and Periodic Method QC Demonstrations: The

following should be conducted as an initial demonstration of
laboratory capability, prior to the analysis of any samples.
Subsequent to this initial demonstration, additional evaluations
of this nature should be conducted on a periodic basis, in
response to changes in instrumentation or operations, and/or in
response to confirmed or suspected systems, method, or
operational problems.
10.3.1
Accuracy and Precision: To demonstrate initial

laboratory capability, analyze a minimum of four
replicate deionized water and clean sand blanks
spiked with the TPH standards in Section 7.4.1 or
7.4.2 at approximately 10 to 20 mg/L (water) and 100
to 200 mg/kg (soil).
227
Appendix
10.3.1.1 Extract and analyze each replicate according

to the procedures described in Section 9.0.
10.3.1.2 Calculate the measured concentrations of
TPH from nC6 to nC35 in all replicates, the
mean accuracy (as a percentage of true
value) for each analyte, and the precision (as
% RSD) of the measurements for each
analyte.
10.3.1.3 For each determination, the mean accuracy,
expressed as a percentage of the true value,
should be between 60% and 140%.
10.3.1.4 If desired, the accuracy and precision
evaluation may be combined with the MDL
evaluation specified in Paragraph 10.3.2.
10.3.2
Method Detection Limits

10.3.2.1 Soil/sediment MDLs are determined by
extracting 7-10 replicates of 10 g of clean
sand blanks which have been fortified with
either of the stock solutions defined in
Sections 7.4.1 or 7.4.2 at approximately
50 mg/kg.
Extract and analyze each
replicate according to the procedures
described in Section 9.0. Calculate the
Method Detection Limit (MDL) using
guidelines in SW -846 [1].
10.3.2.2 Water MDLs are determined by extracting
7-10 replicates of deionized water fortified
with stock solution in 7.4.1 or 7.4.2 at
approximately 5 mg/L. Extract and analyze
each replicate according to the procedure
described in section 9.0. Calculate the MDL
using guidelines in SW-846 [1].
10.3.3
Fractionation: The stock solution described in

Section 7.4.2 can be used to demonstrate the
capability of properly fractionating aliphatic and
aromatic hydrocarbons in a sample.
228
Appendix
10.3.3.1 Prepare the column and follow the

fractionation as described in Section 9.2 for
the fractionation of 1 mL of the stock
solution in Section 7.4.2.
NOTE: The amount of n-pentane used during
fractionation is critical. Excessive n-pentane will
cause elution of aromatics into the aliphatic
fraction. Insufficient n-pentane will cause low
recoveries of the aliphatic fraction. The volume of
n-pentane recommended (8-10 mL) may need to
be adjusted to meet QC limits.
10.3.3.2 For each analyte within the fractionation
check solution, the mean accuracy,
expressed as a percentage of the true value,
should be between 60% and 140%.
10.3.3.3 It is acceptable to encounter a 10 to 20%
crossover of the fractions. This means that
it is within the acceptance criteria of this
method to have 10 to 20% aliphatics in the
aromatic fraction and 10 to 20% aromatics
in the aliphatic fraction.
10.4
Ongoing Method QC Demonstrations

10.4.1
At a minimum, with every batch of 20 samples or

less the lab should analyze the following:
10.4.1.1 Calibration Check Standard: A mid-range
calibration standard, prepared from the same
stock standard solution used to develop the
calibration curve, should be analyzed prior
to sample analysis to verify the calibration
state of the instrument. For large analytical
batches that contain more than 10 samples,
the analysis of an additional mid-range
calibration check standard is recommended
after the analysis of the 10th sample. If the
relative percent difference (RPD) of any
analyte within a calibration check standard
varies from the predicted response by more
than 25%, a new calibration curve should be
prepared for that analyte (see Section 9.5).
229
Appendix
Any sample analyzed after the last

acceptable check standard must be
reanalyzed.
10.4.1.2 Laboratory Control Sample: A soil LCS is
prepared by fortifying 10 g of a clean sand
blank with 0.25 mL to 1.0 mL of one of the
standards described in Sections 7.4.1 and
7.4.2 for spiking solutions. A water LCS is
prepared by fortifying 30 mL of deionized
water with 0.1 mL of standard described in
Sections 7.4.1 and 7.4.2. The spike recovery
should be between lab-generated control
chart limits. If there is insufficient data to
control chart the maximum default, limits of
60% to 140% may be used.
10.4.1.3 Matrix Spike (MS) and Matrix Spike
Duplicates (MSD): A soil matrix spike is
prepared by fortifying an actual sample with
0.25 mL to 1.0 mL of the matrix spiking
solution. A soil matrix spike duplicate is
prepared the same way as the soil matrix
spike. A water matrix spike is prepared by
fortifying an actual sample with 0.1 mL of
the matrix spiking solution. A water matrix
spiked duplicate is prepared the same way as
the water matrix spike. The purpose of the
matrix spike is to determine whether the
sample matrix contributes bias to the
analytical results. The purpose of the matrix
spike duplicate is to determine the precision
of the analysis.
The background
concentrations of the analytes in the sample
matrix should be determined in a separate
aliquot and the measured values in the
matrix
corrected
for
background
concentrations.
The
corrected
concentrations of TPH for the MS spike
should be within the lab-generated control
limits for the LCS. If there is insufficient
data to control chart the maximum default,
limits of 60% to 140% may be used. If the
MS is outside of the control limits, then the
batch it represents should be noted as having
230
Appendix
matrix interference. The LCS should be

used to show the method is not contributing
to spike loss. If the LCS falls outside the
lab-generated control limits then that batch
needs to be reanalyzed until the LCS falls
within the generated control limits. The
RPD of the duplicate samples (MS and
MSD) should not exceed 30%.
If
insufficient sample is available for an MS
and MSD, then an LCS duplicate must be
analyzed. The LCS duplicate recovery must
fall within generated control limits. The
RPD of the LCS and LCS duplicate must be
less than or equal to 30%.
10.4.3
11.0
If any of the performance standards specified in

Section 10.4 are not met, the problem should be
corrected before further samples are analyzed. Any
samples run between the last QC samples that meet
the criteria and those that have fallen out should be
rerun. If this is not possible, that data should be
reported as suspect.
Data Production and Reporting

11.1
Calibration: Using the external calibration procedure

(Section 9.5), calibrate the GC as follows:
11.1.1
Calculate a collective Calibration Factor (CF), or

linear or quadratic regression relationship for the sum
of all the peaks that comprise either of the standards
defined in Sections 7.4.1 or 7.4.2 for the C6 to C35
range or a narrower range if the sample contains a
smaller carbon range and the option is taken to use a
narrower boiling product as a standard. The CF or
regression correlation should be done on the total
area and the total mass of hydrocarbons in the
standard within the specified carbon range.
11.1.2
The CF or regression correlation obtained in Section

11.1.1 can be used to calculate the petroleum
hydrocarbon concentrations for smaller carbon ranges
within the total TPH. These results provide the
approximate boiling point distribution/carbon number
range information. An easier and more convenient
231
Appendix
approach is to calculate the area percents of the

individual sums of the peaks within the individual
carbon markers ranges normalized with respect to the
total area of the C6 to C35 (TPH area). Then use these
percentages to calculate the amounts of petroleum
hydrocarbons within those ranges.
11.2
Data Reporting Format

11.2.1
The following information and data should be

reported:
11.2.1.1 The sample matrix (soil, sediment, sludge).
11.2.1.2 The date(s) the sample was collected,
received by the laboratory, extracted and
analyzed.
11.2.1.3 Note in the report if there were any
problems observed with the samples as
received, such as the physical condition of
the containers, the temperature of the
samples, and the use of appropriate
preservatives. No need to inclu de this
information if no problems observed.
11.2.1.4 Moisture content if desired (not required in
this method).
11.2.1.5 The calculated concentrations of TPH C6 to
C35 (or whatever carbon range the sample
contains),
the
approximate
boiling
point/carbon number distribution for the
TPH.
11.2.1.6 If sample extract is fractionated, the
calculated concentrations of aliphatic and of
aromatic hydrocarbons C6 to C35 (or
whatever carbon range the sample contains),
the approximate boiling point/carbon
number distribution for the fractions.
11.2.1.7 The method reporting limit for the TPH,
aliphatic, and aromatic hydrocarbons as well
as for the narrower ranges.
232
Appendix
11.2.1.8 Chromatograms and data tables.
12.0
13.0
Method Performance
12.1
The method has been applied to the analysis of neat crude oil,
gasoline, JP-4, and diesel. In addition, the method has been
used for the analysis of soil samples impacted with crude oil
and with petroleum products with different degrees of
weathering. Recoveries are typically 80% or better for most
samples.
12.2
A previous version of this method for TPH and approximate

boiling point/carbon number distribution was tested by 12
laboratories. Single operator precision, overall precision and
method accuracy were determined and found to be 13%, 30%,
and 84%, respectively. A similar study performed on this
approach and the fractionation procedure by a single analyst
showed an average accuracy of 80% with an average overall
RSD of 6%. Also, an independent laboratory evaluation of
this method resulted in a single analyst average accuracy of
111% and an overall RSD of 10%.
12.3
Additional method refinement and evaluation is in progress.
Pollution Prevention
13.1
The solvent used in this method poses little threat ot the

environment when recycled and managed properly.
13.2
The quantity of chemicals purchased should be based on the

expected usage during its shelf life. Standards should be
prepared in volumes consistent with laboratory use to
minimize the volume of expired standards to be disposed.
233
Appendix
14.0
Waste Management
14.1
15.0
It is the laboratory's responsibility to comply with all federal,

state, and local regulations governing waste management,
particularly to comply with the hazardous waste identification
rules and land disposal restrictions, and to protect the air,
water, and land by minimizing and controlling all releases
from fume hoods and bench operations. Compliance with all
sewage discharge permits and regulations is also required.
References
1.
USEPA, Test Method for the Evaluation of Solid Waste (SW846), 3rd Edition (1996).
2.
Tong, H.Y., and Karasek, F.W., Flame Ionization Detector

Response Factors for Compound Classes in Quantitative Analysis
of Complex Organic Mixtures, Analytical Chemistry, 56, 21242128 (1984).
3.
TNRCC TX Method 1005, Total Petroleum Hydrocarbons,

December 1997.
4.
API Publication No. 4599, Interlaboratory Study of Three

Methods for Analyzing Petroleum Hydrocarbons in Soils: DieselRange Organics (DRO), Gasoline-Range Organics (GRO),
Petroleum Hydrocarbons (PHC), June 1994.
5.
American Association of Railroads, Risk Management Division,

Comparison of Analytical Methods for Use in Evaluating the Risk
From Petroleum Hydrocarbons in Soil, prepared by Remediation
Technologies, Inc., R-903 (1997).
6.
Total Petroleum Hydrocarbons Criteria Working Group Series,

Volume 3, Selection of Representative TPH Fractions Based on
Fate and Transport Considerations, Amherst Scientific
Publishers , July 1997.
7.
USEPA Test Method for the Evaluation of Solid Waste (SW846), Method 3611B, Alumina Column Cleanup and Separation
of Petroleum Wastes (1996).
8.
USEPA Test Method for the Evaluation of Solid Waste (SW846), Method 3630C, Silica Gel Cleanup (1996).
9.
Massachusetts Department of Environmental Protection, Method

for the Determination of Extractable Petroleum Hydrocarbons
(EPH) (1995).
234
Appendix
10. Washington State WTPH-HCID Method.

11. Rhodes, I.A.L., Olvera, R.Z., Leon, J.L., and Hinojosa, E.M.,
Determination of Total Petroleum Hydrocarbons by Capillary
Gas Chromatography, Proceedings of the Fourteenth Annual
EPA Conference on Analysis of Pollutants in the Environment,
Norfolk, Virginia (1991).
12. Rhodes, I.A.L, Hinojosa, E.M., Barker, D.A., and Poole, R.L,
Pitfalls Using Conventional TPH Methods for Source
Identification: A Case Study, Proceedings of the Fourteenth
Annual EPA Conference on Analysis of Pollutants in the
Environment, Norfolk, Virginia (1994).
13. Rhodes, I.A.L, Hinojosa, E.M., Barker, D.A., and Poole, R.L,
Conventional
TPH
Pitfalls,
Environmental
Lab,
December/January (1995/96).
14. Dempster, H.S., Lollar, B.S., and Feenstra, S., Tracing Organic
Contaminants in Groundwater: A New Methodology Using
Compound-Specific Isotopic Analysis, Environmental Science
and Technology, 31, 3193-3197 (1997).
15. ASTM Method D 4547-91, Standard Practice for Sampling
Waste and Soils for Volatile Organics (1991).
235
Glossary of Terms
Glossary of Terms
Acute: Rapid. An acute exposure is of short duration.
Aliphatic: Organic compounds which are non-aromatic and non-polar, such as
alkanes and alkenes.
Aliquot: A precise portion.
Alkane: A saturated hydrocarbon, containing only hydrogen and carbon in
single bonds. Synonymous with saturate fraction, paraffins, and naphthenes.
Alkene: Hydrocarbons containing carbon-carbon double bonds. Alkenes are
refinery by-products of the cracking of oils.
Aromatic: Unsaturated (double-bond) cyclic hydrocarbons containing the
benzene ring as the basic structural unit of the molecule.
API Gravity: The American Petroleum Institute method for specifying the
specific gravity of crude oil.
Benzene Ring: The simplest aromatic compound, containing one, 6-membered
carbon ring with double and single carbon-carbon bonds.
Biochemical: Pertaining to the chemistry of biological processes.
Biophysical: Pertaining to the physical nature of biological processes.
Cancer Slope Factor: A conservative estimate of the incremental probability of
a person contracting cancer from a unit intake of a chemical over a lifetime.
Carcinogen: A chemical, biologic, or physical agent that can cause cancer.
Chemical of Potential Concern: A chemical that has the potential to negatively
impact human health and the environment at a site.
Chronic: Occurring over a long period. Chronic exposures generally occur over
a long period of time. USEPA defines chronic exposures as those that may last
at least 10% of an average lifetime (i.e., 7 years or more).
Comprehensive Environmental Response and Contingency Liability Act
(CERCLA): 1980 United States federal law authorizing identification and
remediation of unsupervised hazardous waste sites and spill reporting (also
called Superfund).
Detection Limit: The minimum concentration of a chemical that can be
detected using a specific measurement procedure and laboratory equipment.
Dose-Response: The measurement of the response of an organism to a dose of
chemical.
Eluent: The mobile liquid phase used to separate chemical fractions of oil
during liquid chromatography.
Exposure: The means by which receptors come in contact with chemicals.
236
Glossary of Terms
Exposure Assessment: A three-step process in which the exposure setting is

characterized, complete exposure pathways are identified, and the magnitude of
the potential exposure is estimated.
Exposure Pathway: The path that a chemical takes from its source to a
receptor.
Fingerprint: A chromatographic signature. Can be used for qualitative or
quantitative comparisons of the composition of oils.
Fractionation, TPH: The separation of the hydrocarbon extract into compound
classes such as aliphatic and aromatic hydrocarbons.
Gas Chromatography: The separation of mixtures of compounds by partition
between a mobile gas phase and a stationary liquid phase.
Granuloma: A growth or tumor that appears granulated.
Hazard Identification: A review of site assessment data and the identification
of chemicals that may be present at the site and may be of concern.
Hazard Quotient: The ratio of an estimated long-term daily exposure level to
an acceptable exposure level. A Hazard Quotient of 1 or less is deemed to be
acceptable by the United States Environmental Protection Agency.
Hematological: Pertaining to the blood system.
Hepatotoxicity: Liver toxicity.
Hepatic: Pertaining to the liver.
Indigenous: Naturally-occurring.
Intervention Value: As defined by the Dutch government, the concentration
above which soils are considered to be seriously contaminated.
Lacustrine: Pertaining to, produced by, or formed in a lake. Some crude oils
originate from lacustrine source rocks.
Leaching Factor: A factor that will predict the amount of contaminant that will
partition between soil and water.
Nephrotoxicity: Kidney toxicity.
Non-Aqueous Phase Liquid: A liquid other than water that exists in the
subsurface environment.
Non-Carcinogen: A chemical that is not expected to cause cancer, but may
cause other adverse health effects.
Oil and Grease: Measurement of the amount of relatively non-volatile
hydrocarbons (as well as vegetable oils, animal fats, waxes soaps, greases, and
related matter). An aqueous or solid sample measured by gravimetric analysis
of a solvent extract.
Oral Cancer Slope Factor: See Cancer Slope Factor.
Pathway: See Exposure Pathway.
237
Glossary of Terms
Phytoavailable: Available for uptake into the tissues of plants.

Phytotoxic: Adverse effects of chemicals on plants.
Polar Compound: An organic compound with distinct regions of partial
positive and negative charge. Polar compounds include alcohols and aromatics.
Polyaromatic Hydrocarbon (Polycyclic Aromatic Hydrocarbons): A
hydrocarbon that contains more than one benzene ring.
Polynuclear Aromatic: See Polyaromatic Hydrocarbon.
Preliminary Remediation Goal: As defined by the United States
Environmental Protection Agency Region IX, a risk-based concentration of a
chemical in an environmental medium (e.g., soil, water, or air) that is considered
to be protective of human health and that can be used as a screening level for
evaluating contaminant levels at individual sites, as appropriate.
Protective Concentration Level: As defined by the Texas Natural Resources
Conservation Commission, the concentration of a chemical of concern which
can remain within the source medium and not result in levels which exceed the
applicable human health risk-based exposure limit or ecological protective
concentration level at the point of exposure for that exposure pathway.
Receptor: A human being or other organism that has the potential to be exposed
to chemicals.
Reference Concentration: An estimate of continuous inhalation exposure to a
chemical that is likely to be without an appreciable risk of adverse effect over a
lifetime.
Reference Dose: An estimate of daily exposure to a chemical that is likely to be
without an appreciable risk of adverse effect over a lifetime.
Residual: The presence of a free product or oil in soil; also called separatephase hydrocarbons.
Resource Conservation and Recovery Act (RCRA): Legislation that was
enacted in the United States in 1976 to regulate solid waste and resource
recovery for all but a few exempt wastes .
Risk-Based Corrective Action: A tiered decision-making approach for site
assessment, risk assessment, and site management.
Risk-Based Decision-Making: A decision-making approach to site
management based on an evaluation of the risks that a given situation might
present.
Risk-Based Screening Levels: Chemical-specific concentrations in
environmental media that are considered to be protective of human health and
the environment.
Risk Characterization: The final step of a risk evaluation, which combines the
results of the exposure and toxicity assessments in order to quantify the potential
risks to human health and the environment.
238
Glossary of Terms
Saturated Hydrocarbon: A hydrocarbon that includes normal and branched

alkanes and cycloalkanes (paraffins and naphthenes). Saturates are the nonaromatic hydrocarbon fraction of an oil.
Site Specific Target Level (SSTL): As defined by the American Society for
Testing and Materials, a risk-based remedial action target level for chemical(s)
of concern developed for a particular site under the Tier 2 and Tier 3
evaluations.
Soil Screening Level: As developed by the United States Environmental
Protection Agency, a risk-based concentration of a chemical in soil that is
protective for potential human health impacts.
Target Value: As defined by the Dutch government, the concentration of a
chemical in soil that is considered to be desirable and is generally based on
background concentrations of chemicals in soil in the Netherlands.
Total Organic Carbon: The quantity of organic carbon in a solid or aqueous
sample as measured by wet oxidation of the sample.
Total Petroleum Hydrocarbon: The quantity of extractable compounds
detected in a sample of soil or water as measured by the detection methods (GC,
infrared, gravimetric) in a solvent ext ract of soil or water.
Toxicology: The study of the harmful effects of chemicals on organisms.
Toxicity Assessment: An evaluation of a chemical for its ability to cause cancer
or other adverse health effects.
Unsaturated Hydrocarbon: A hydrocarbon containing one or more double or
triple bonds, such as alkenes.
Unsaturated Soil: A soil having pore spaces that are not completely filled with
water or other fluids.
Volatilization Factor: A factor that predicts the amount of contaminant that
will partition between the soil and air.
239

Archive RiskBasedBook

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Archive RiskBasedBook

Uploaded by

Copyright:

Available Formats

Risk-Based Decision-Making for

Assessing Petroleum Impacts at

RISK-BASED DECISION-MAKING FOR ASSESSING

Risk-Based Decision-Making for

THIS BOOK WAS PUBLISHED IN COOPERATION WITH

Risk-Based Decision-Making for

Recognizes the differences of each site,

In terms of risk-based assessments, for sites with contaminated soils, some of

The fact that a chemical, such as a petroleum hydrocarbon, can be

Not all chemicals in a soil are easily released and leachable.

It is possible to allow some chemicals to remain at a site and still be

Reasonable information is available to evaluate site-specific risks of

It is not mass removal of chemicals at a site, but rather risk reduction

Some concentrations of anthropogenic chemicals in a soil may have

The site remediation question is not how much contamination can be

Information about the chemical release in a soil and chemical exposure

Natural attenuation of chemicals such as petroleum hydrocarbons does

Aging and weathering of anthropogenic hydrocarbons in soil can result

Detailed knowledge on many of these and other important points is incorporated

decisions for the management of hydrocarbon-impacted soil at oil and gas

Rebecca L. Carovillano, CH2M HILL

In addition we would like to acknowledge the following individuals for their

Rene F. Bernier, Texaco Incorporated

Groundwater mixing zone thickness

Degradation constant in the saturated zone

Degradation constant in the unsaturated zone beneath the

Volatilization rate constant for the contaminant source zone

Soil bulk density

Soil particle density

Air filled porosity

Volumetric air content

Volumetric oil content in vadose zone soil

Porosity filled with residual phase

Water filled porosity

Volumetric water content

Cross-sectional area of the source

Average dose level for an exposure pathway

British Columbia Environment

Benzene, toluene, ethylbenzene, xylenes

State of California Environmental Protection Agency

Comprehensive Environmental Response and Contingency

Concentration in organic phase

Final concentration of chemical in the oil phase

Concentration in vapor reported as parts per million volume

Concentration of chemical sorbed to soil

Conceptual site model

Total concentration of chemical in soil

Concentration of chemical in vapor phase

Concentration of chemical in water

Maximum dissolved concentration

Cold-vapor atomic absorption spectrometry

Dilution attenuation factor in groundwater

Dilution attenuation factor in the groundwater mixing zone

Dilution attenuation factor in groundwater downstream of the

Dilution attenuation factor in unsaturated zone

Molecular diffusion coefficient for chemical in air

Molecular diffusion coefficient for chemical in water

Total effective diffusion coefficient

Diesel range organics

Electronic pressure control

Fraction of oil in soil

Gas chromatograph or gas chromatography

Graphite furnace atomic absorption spectrometry

Henrys Law constant or distance from bottom of contaminant